ANSI/AAMI EC38:2007

Medical electrical equipment –
Part 2-47: Particular requirements
for the safety, including essential
performance, of ambulatory
electrocardiographic systems

Association for the Advancement
of Medical Instrumentation
Single user license only. Copying, networking, and distribution prohibited.

Objectives and uses of AAMI standards and
recommended practices
It is most important that the objectives and potential uses of an
AAMI product standard or recommended practice are clearly
understood. The objectives of AAMI's technical development
program derive from AAMI's overall mission: the advancement of
medical instrumentation. Essential to such advancement are (1) a
continued increase in the safe and effective application of current
technologies to patient care, and (2) the encouragement of new
technologies. It is AAMI's view that standards and recommended
practices can contribute significantly to the advancement of
medical instrumentation, provided that they are drafted with
attention to these objectives and provided that arbitrary and
restrictive uses are avoided.
A voluntary standard for a medical device recommends to the
manufacturer the information that should be provided with or on
the product, basic safety and performance criteria that should be
considered in qualifying the device for clinical use, and the
measurement techniques that can be used to determine whether the
device conforms with the safety and performance criteria and/or to
compare the performance characteristics of different products.
Some standards emphasize the information that should be provided
with the device, including performance characteristics, instructions
for use, warnings and precautions, and other data considered
important in ensuring the safe and effective use of the device in the
clinical environment. Recommending the disclosure of
performance characteristics often necessitates the development of
specialized test methods to facilitate uniformity in reporting;
reaching consensus on these tests can represent a considerable part
of committee work. When a drafting committee determines that
clinical concerns warrant the establishment of minimum safety and
performance criteria, referee tests must be provided and the reasons
for establishing the criteria must be documented in the rationale.
A recommended practice provides guidelines for the use, care,
and/or processing of a medical device or system. A recommended
practice does not address device performance per se, but rather
procedures and practices that will help ensure that a device is used
safely and effectively and that its performance will be maintained.
Although a device standard is primarily directed to the
manufacturer, it may also be of value to the potential purchaser or
user of the device as a frame of reference for device evaluation.
Similarly, even though a recommended practice is usually oriented
towards healthcare professionals, it may be useful to the
manufacturer in better understanding the environment in which a
medical device will be used. Also, some recommended practices,
while not addressing device performance criteria, provide
guidelines to industrial personnel on such subjects as sterilization
processing, methods of collecting data to establish safety and
efficacy, human engineering, and other processing or evaluation
techniques; such guidelines may be useful to health care
professionals in understanding industrial practices.
In determining whether an AAMI standard or recommended
practice is relevant to the specific needs of a potential user of the
document, several important concepts must be recognized:
All AAMI standards and recommended practices are voluntary
(unless, of course, they are adopted by government regulatory or
procurement authorities). The application of a standard or
recommended practice is solely within the discretion and
professional judgment of the user of the document.

Single user license only. Copying, networking, and distribution prohibited.

Each AAMI standard or recommended practice reflects the
collective expertise of a committee of health care professionals and
industrial representatives, whose work has been reviewed
nationally (and sometimes internationally). As such, the consensus
recommendations embodied in a standard or recommended practice
are intended to respond to clinical needs and, ultimately, to help
ensure patient safety. A standard or recommended practice is
limited, however, in the sense that it responds generally to
perceived risks and conditions that may not always be relevant to
specific situations. A standard or recommended practice is an
important reference in responsible decision-making, but it should
never replace responsible decision-making.
Despite periodic review and revision (at least once every five
years), a standard or recommended practice is necessarily a static
document applied to a dynamic technology. Therefore, a standards
user must carefully review the reasons why the document was
initially developed and the specific rationale for each of its
provisions. This review will reveal whether the document remains
relevant to the specific needs of the user.
Particular care should be taken in applying a product standard
to existing devices and equipment, and in applying a recommended
practice to current procedures and practices. While observed or
potential risks with existing equipment typically form the basis for
the safety and performance criteria defined in a standard,
professional judgment must be used in applying these criteria to
existing equipment. No single source of information will serve to
identify a particular product as "unsafe". A voluntary standard can
be used as one resource, but the ultimate decision as to product
safety and efficacy must take into account the specifics of its
utilization and, of course, cost-benefit considerations. Similarly, a
recommended practice should be analyzed in the context of the
specific needs and resources of the individual institution or firm.
Again, the rationale accompanying each AAMI standard and
recommended practice is an excellent guide to the reasoning and
data underlying its provision.
In summary, a standard or recommended practice is truly
useful only when it is used in conjunction with other sources of
information and policy guidance and in the context of professional
experience and judgment.

INTERPRETATIONS OF AAMI STANDARDS
AND RECOMMENDED PRACTICES
Requests for interpretations of AAMI standards and recommended
practices must be made in writing, to the AAMI Vice President,
Standards Policy and Programs. An official interpretation must be
approved by letter ballot of the originating committee and
subsequently reviewed and approved by the AAMI Standards
Board. The interpretation will become official and representation of
the Association only upon exhaustion of any appeals and upon
publication of notice of interpretation in the "Standards Monitor"
section of the AAMI News. The Association for the Advancement
of Medical Instrumentation disclaims responsibility for any
characterization or explanation of a standard or recommended
practice which has not been developed and communicated in
accordance with this procedure and which is not published, by
appropriate notice, as an official interpretation in the AAMI News.

American National Standard

ANSI/AAMI EC38:2007
(Revision of ANSI/AAMI EC38:1998)

Medical electrical equipment – Part 2-47:
Particular requirements for the safety, including
essential performance, of ambulatory
electrocardiographic systems

Developed by
Association for the Advancement of Medical Instrumentation
Approved 10 December 2007 by
American National Standards Institute, Inc.

Abstract:

To establish particular requirements for the safety, including essential performance, of
ambulatory electrocardiographic systems that provide continuous recording and analysis
of ECG.

Keywords:

ambulatory electrocardiographic system, ambulatory recorder, electrocardiogram, electrode,
continuous recorder

Single user license only. Copying, networking, and distribution prohibited.

marketing.org © 2008 by the Association for the Advancement of Medical Instrumentation All Rights Reserved Publication. including civil and criminal penalties. voluntary technical documents are adopted by government regulatory agencies or procurement authorities. electronically or otherwise. For permission regarding the use of all or any part of this document. AAMI procedures require that action be taken to reaffirm. et seq. technical information reports. The existence of an AAMI standard does not in any respect preclude anyone. Glebe Road. purchasing. and users are cautioned to obtain the latest editions. Printed in the United States of America ISBN 1-57020-312-1 Single user license only. or procedures not conforming to the standard. Phone: (703) 525-4890. contact AAMI at 1110 N. or using products. in which case the adopting agency is responsible for enforcement of its rules and regulations. and distribution prohibited. § 101. Interested parties may obtain current information on all AAMI standards by calling or writing AAMI. Glebe Road. and damages of $100. Suite 220.C. networking. recommended practices. Copying. All AAMI standards. and other types of technical documents developed by AAMI are voluntary.000 per offense.) to make copies of all or any part of this document (whether internally or externally) without the prior written permission of the Association for the Advancement of Medical Instrumentation. or withdraw this standard no later than five years from the date of publication. Published by Association for the Advancement of Medical Instrumentation 1110 N. AAMI standards are subject to periodic review. from manufacturing. VA 22201-4795. CAUTION NOTICE: This AAMI standard may be revised or withdrawn at any time. of all or any part of this document without the prior written permission of the Association for the Advancement of Medical Instrumentation is strictly prohibited by law.S. whether they have approved the standard or not.aami. or transmission. Violators risk legal action. Suite 220 Arlington. and their application is solely within the discretion and professional judgment of the user of the document. reproduction. Fax: (703) 525-1067. revise. Occasionally.AAMI Standard This Association for the Advancement of Medical Instrumentation (AAMI) standard implies a consensus of those substantially concerned with its scope and provisions. VA 22201-4795 www. It is illegal under federal law (17 U. photocopying. Arlington. storage. processes. .

...................................................................................................................................................................... 1 1..................................................................1 Scope ............................................... 7 36............. 4 SECTION TWO – ENVIRONMENTAL CONDITIONS *10 Environmental conditions .............................................................................................. vii Committee representation ..............................3 Values of test voltages .............................. 1 1................................................................................. xi Foreword ............................................................. networking...................................................................... 5 SECTION FOUR – PROTECTION AGAINST MECHANICAL HAZARDS 21 Mechanical strength ..1 Environment ....................................................................5 Collateral Standards.............................................201 Emissions ........................... xxix SECTION ONE – GENERAL 1 Scope and object .................... 7 36..................................202 Immunity...........................2 Object ................... marking and documentation .....2 EQUIPMENT or EQUIPMENT parts .............................................................................................................................................................2 Radiated radio-frequency electromagnetic fields .............. ......2............................................................... 5 20.....2 Particular requirements for EQUIPMENT with an APPLIED PART ........................................... x AAMI Deviations to IEC 60601-2-47:2001 ......................................1 Radio frequency (RF) emissions....................... 5 10...... 8 SECTION SIX – PROTECTION AGAINST HAZARDS OF IGNITION OF FLAMMABLE ANAESTHETIC MIXTURES SECTION SEVEN – PROTECTION AGAINST EXCESSIVE TEMPERATURES AND OTHER SAFETY HAZARDS Single user license only.....6 Magnetic fields............ 2 5 Classification .............1 Electrostatic discharge ...........................................202.... 7 36.............................................................................................3 Particular Standards................................................................................................................................. and distribution prohibited.............................. 5 20..........201. 6 SECTION FIVE – PROTECTION AGAINST HAZARDS FROM UNWANTED OR EXCESSIVE RADIATION *36 ELECTROMAGNETIC COMPATIBILITY....................Contents Page Glossary of equivalent standards ...8......................................................................................................202.................................................................................................... 3 6 Indentation............................................................................................................................................... Copying........ 1 1............................ 1 1...........................................................202............................................ 5 SECTION THREE – PROTECTION AGAINST ELECTRIC SHOCK HAZARDS 20 Dielectric strength ........................................................................................................................... 3 6................. 4 Instructions for use ...................................1 Marking on the outside of 6.................................................................. ix Background of ANSI/AAMI adoption of IEC 60601-2-47:2001 ........................ 7 36.... 2 *2 Terminology and definitions.................. 7 *36.......... 7 *36.............................. xxvii Introduction ......................................................................................................................................................

.......13 Hard copy grid standard ......................... 19 *56........7........................................................... 11 *51.............15 Temporal alignment.........................................................................................................................................19 Patient event marks .....3 Physician report .........5......................................9 Frequency response................ 15 *51...14 Gain settings and switching .................5.................................................................................. 18 51................................................................................................................... 19 SECTION NINE – ABNORMAL OPERATION AND FAULT CONDITIONS...........................................5.......................................5...7 Batteries ..................................5...8 Multichannel crosstalk.................. 12 *51...... 16 *51................5. 17 *51.5........................................................................1 Standard databases to be used to evaluate automated analyses.............................................. 17 51...............minimum requirements ...101.........................................5.......SECTION EIGHT – ACCURACY OF OPERATING DATA AND PROTECTION AGAINST HAZARDOUS OUTPUT 50 Accuracy of operating data ........... 13 *51.... 9 *50............... 16 *51.. networking........................................................................18 Baseline stability ...5......12 Timing accuracy ..................... 10 *51... 12 *51.............................................................2 Input impedance .........................................6 Amplitude calibration .......... 19 51.............5...........11 Function in the presence of pacemaker pulse .................................................... 9 51 Protection against hazardous output............. and distribution prohibited..............................................................101 Automated analysis..........5 Gain stability ...........................4 Gain accuracy ................................. 13 *51..........10 Minimum feature size ......................5...................................................................... 19 51................................16 Special considerations for high speed superimposition display (optional)......................................................................101. 19 *56................................................................................. 10 *51. ENVIRONMENTAL TESTS SECTION TEN – CONSTRUCTIONAL REQUIREMENTS *56 Components and general assembly .5..5.......................101 Monitoring time and retention of data ................................................................................................................................. 14 *51............ 9 *50................... 13 *51...............3 Common mode rejection...5....5................. 11 *51.... 8 *50................. ..........2 Performance reporting requirements .......................5.7 System noise . 10 *51..........................5............. 12 *51.................20 Annexes Appendix L—References – Publications mentioned in this Standard...........101...............................................5 Incorrect output...............................5..... 4 Single user license only....................1 Input dynamic range................................................................17 Lead Definitions ....................5..................................20 Full disclosure (miniature displays)............................................. Copying......... 9 *50.............5........................................................ 17 51......................... 26 Tables Table 101—Lead color codes . 25 Annex AA (informative)—Guidance and rationale ..............................................

...... ........202........ 23 Figure 106—Test circuit for pacemaker pulse tolerance according to 51................. 21 Figure 103—Test signal for input dynamic range test according to 51........................ 20 Figure 102—Test set-up for radiated emission and radiated immunity test according to 36.............2 .. networking.........5.......... and distribution prohibited......................5..........................1........................201.Figures Figure 101—Test set-up for conductive emission test according to 36........5.... 22 Figure 105—Test circuit for common mode rejection according to 51..................1 and 36......................201..3............... Copying.... 22 Figure 104—General test circuit for 51...............11 .............. 24 Single user license only.........1 ..5...........

Single user license only. . networking. and distribution prohibited. Copying.

adoption by AAMI.S. NOTE: Documents are sorted by international designation. For each International Standard that has been adopted by AAMI (and ANSI). and distribution prohibited.S. designation and level of equivalency to the International Standard. International designation U. Copying. vii . networking. the table below gives the corresponding U.Glossary of equivalent standards International Standards adopted in the United States may include normative references to other International Standards. Other normatively referenced International Standards may be under consideration for U. this list should not be considered exhaustive. designation Equivalency IEC 60601-1:2005 IEC 60601-1-2:2007 IEC 60601-2-2:2006 IEC 60601-2-4:2002 IEC 60601-2-19:1990 and Amendment 1:1996 IEC 60601-2-20:1990 and Amendment 1:1996 IEC 60601-2-21:1994 and Amendment 1:1996 IEC 60601-2-24:1998 IEC 60601-2-47:2001 IEC 60601-2-50:2001 IEC/TR 60878:2003 IEC/TR 62296:2003 IEC 62304:2006 IEC/TR 62348:2006 ISO 5840:2005 ISO 7198:1998 ISO 7199:1996 ISO 8637:2004 ISO 8638:2004 ISO 10993-1:2003 ISO 10993-2:2006 ISO 10993-3:2003 ISO 10993-4:2002 and Amendment 1:2006 ISO 10993-5:1999 ISO 10993-6:2007 ISO 10993-7:1995 ISO 10993-9:1999 ISO 10993-10:2002 and Amendment 1:2006 ANSI/AAMI ES60601-1:2005 ANSI/AAMI/IEC 60601-1-2:2007 ANSI/AAMI/IEC 60601-2-2:2006 ANSI/AAMI DF80:2003 ANSI/AAMI II36:2004 Major technical variations Identical Identical Major technical variations Major technical variations ANSI/AAMI II51:2004 Major technical variations ANSI/AAMI/IEC 60601-2-21 and Amendment 1:2000 (consolidated texts) ANSI/AAMI ID26:2004 ANSI/AAMI EC38:2007 ANSI/AAMI/IEC 60601-2-50:2006 ANSI/AAMI/IEC TIR60878:2003 ANSI/AAMI/IEC TIR62296:2003 ANSI/AAMI/IEC 62304:2006 ANSI/AAMI/IEC TIR62348:2006 ANSI/AAMI/ISO 5840:2005 ANSI/AAMI/ISO 7198:1998/2001/(R)2004 ANSI/AAMI/ISO 7199:1996/(R)2002 ANSI/AAMI RD16:2007 ANSI/AAMI RD17:2007 ANSI/AAMI/ISO 10993-1:2003 ANSI/AAMI/ISO 10993-2:2006 ANSI/AAMI/ISO 10993-3:2003 ANSI/AAMI/ISO 10993-4:2002 and Amendment 1:2006 ANSI/AAMI/ISO 10993-5:1999 ANSI/AAMI/ISO 10993-6:2007 ANSI/AAMI/ISO 10993-7:1995/(R)2001 ANSI/AAMI/ISO 10993-9:1999/(R)2005 ANSI/AAMI BE78:2002 ANSI/AAMI BE78:2002/A1:2006 ANSI/AAMI/ISO 10993-11:2006 ANSI/AAMI/ISO 10993-12:2007 ANSI/AAMI/ISO 10993-13:1999/(R)2004 ANSI/AAMI/ISO 10993-14:2001/(R)2006 ANSI/AAMI/ISO 10993-15:2000/(R)2006 ANSI/AAMI/ISO 10993-16:1997/(R)2003 ANSI/AAMI/ISO 10993-17:2002 Identical ISO 10993-11:2006 ISO 10993-12:2007 ISO 10993-13:1998 ISO 10993-14:2001 ISO 10993-15:2000 ISO 10993-16:1997 ISO 10993-17:2002 Major technical variations Major technical variations Identical Identical Identical Identical Identical Identical Identical Identical Major technical variations Major technical variations Identical Identical Identical Identical Identical Identical Identical Identical Minor technical variations Identical Identical Identical Identical Identical Identical Identical Identical © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.S. therefore.

networking. Copying. designation Equivalency ISO 10993-18:2005 ISO/TS 10993-19:2006 ISO/TS 10993-20:2006 ISO 11135-1:2007 ISO 11137-1:2006 ISO 11137-2:2006 (2006-08-01 corrected version) ISO 11137-3:2006 ISO 11138-1: 2006 ISO 11138-2: 2006 ISO 11138-3: 2006 ISO 11138-4: 2006 ISO 11138-5: 2006 ISO/TS 11139:2006 ISO 11140-1:2005 ISO 11140-3:2007 ISO 11140-4:2007 ISO 11140-5:2007 ISO 11607-1:2006 ISO 11607-2:2006 ISO 11737-1: 2006 ISO 11737-2:1998 ISO 11737-3:2004 ISO 13485:2003 ISO 14155-1:2003 ISO 14155-2:2003 ISO 14160:1998 ISO 14161:2000 ISO 14937:2000 ANSI/AAMI BE83:2006 ANSI/AAMI/ISO TIR10993-19:2006 ANSI/AAMI/ISO TIR10993-20:2006 ANSI/AAMI/ISO 11135-1:2007 ANSI/AAMI/ISO 11137-1:2006 ANSI/AAMI/ISO 11137-2:2006 Major technical variations Identical Identical Identical Identical Identical ANSI/AAMI/ISO 11137-3:2006 ANSI/AAMI/ISO 11138-1:2006 ANSI/AAMI/ISO 11138-2:2006 ANSI/AAMI/ISO 11138-3:2006 ANSI/AAMI/ISO 11138-4:2006 ANSI/AAMI/ISO 11138-5:2006 ANSI/AAMI/ISO 11139:2006 ANSI/AAMI/ISO 11140-1:2005 ANSI/AAMI/ISO 11140-3:2007 ANSI/AAMI/ISO 11140-4:2007 ANSI/AAMI/ISO 11140-5:2007 ANSI/AAMI/ISO 11607-1:2006 ANSI/AAMI/ISO 11607-2:2006 ANSI/AAMI/ISO 11737-1:2006 ANSI/AAMI/ISO 11737-2:1998 ANSI/AAMI/ISO 11737-3:2004 ANSI/AAMI/ISO 13485:2003 ANSI/AAMI/ISO 14155-1:2003 ANSI/AAMI/ISO 14155-2:2003 ANSI/AAMI/ISO 14160:1998 ANSI/AAMI/ISO 14161:2000 ANSI/AAMI/ISO 14937:2000 Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical Identical ISO/TR 14969:2004 ANSI/AAMI/ISO TIR14969:2004 Identical ISO 14971:2007 ISO 15223-1:2007 ISO 15225:2000 and A1:2004 ANSI/AAMI/ISO 14971:2007 ANSI/AAMI/ISO 15223-1:2007 ANSI/AAMI/ISO 15225:2000/(R)2006 and A1:2004/(R)2006 ANSI/AAMI/ISO 15674:2001 ANSI/AAMI/ISO 15675:2001 ANSI/AAMI/ISO 15882:2003 ANSI/AAMI/ISO TIR16142:2005 ANSI/AAMI ST81:2004 ANSI/AAMI/ISO 17665-1:2006 ANSI/AAMI/ISO 18472:2006 ANSI/AAMI/ISO 19218:2005 ANSI/AAMI/ISO 22442-1:2007 ANSI/AAMI/ISO 22442-2:2007 ANSI/AAMI/ISO 22442-3:2007 ANSI/AAMI/ISO 25539-1:2003 and A1:2005 ANSI/AAMI/ISO 81060-1:2007 Identical Identical Identical ISO 15674:2001 ISO 15675:2001 ISO 15882:2003 ISO/TR 16142:2006 ISO 17664:2004 ISO 17665-1:2006 ISO 18472:2006 ISO/TS 19218:2005 ISO 22442-1:2007 ISO 22442-2:2007 ISO 22442-3:2007 ISO 25539-1:2003 and A1:2005 ISO 81060-1:2007 viii Identical Identical Identical Identical Major technical variations Identical Identical Identical Identical Identical Identical Identical Identical © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.International designation U.S. . and distribution prohibited.

Welch Allyn. ix . Draeger Medical Kay Rutishauser. Luke’s Roosevelt Hospital Richard A. U. Sunderland. Massachusetts Institute of Technology William Murray. U. Committee approval of the standard does not necessarily imply that all committee members voted for its approval.E. Alaris Medical Systems Inc. Pantiskas Robert E. Food and Drug Administration George Moody. Welch Allyn. Medicomp. Inc. Jr. networking. the AAMI Electrocardiograph Committee had the following members: Cochairs: Members: Carole Carey Carl A. Bruce. Medical Systems Benjamin C. Medicomp. Food and Drug Administration Stephen A. Carole Carey. Marinello. American Association of Critical Care Nurses Ahmad Sajadi. Philips Medical At the time this document was published. Philips Medical Systems NOTE—Participation by federal agency representatives in the development of this standard does not constitute endorsement by the federal government or any of its agencies.Committee representation Association for the Advancement of Medical Instrumentation Electrocardiograph (ECG) Committee This standard was developed by the ECG/Ambulatory Electrocardiograph Working Group of the Electrocardiograph Committee of the Association for the Advancement of Medical Instrumentation.S. Brian J. G. California Richard Gregg. Fresno. Sophia Zhou. Copying. St. the committee’s ECG/Ambulatory Electrocardiograph Working Group had the following members: Cochairs: Members: Alternates: Robert E. Healthcare John Wang. Massachusetts Institute of Technology Carl A. MD. Inc. Eloff. Sunderland. Deedwania. ConMed Corp. At the time this document was published. Inc. Food and Drug Administration Arthur R.. Young. Pantiskas. RN. Philips Medical Charles Ho.S. G. Inc. and distribution prohibited.E. George Moody. Richard A.S. U. PhD. Draeger Medical Mary Schneider. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. Eddy. VA Medical Center. MD. Draeger Medical Jonathan Steinberg. Bruce George Moody Robert E. Prakash C. Bruce.

was responsible for developing U. x © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. The standard will also be useful to ambulatory monitoring equipment users because knowledge of the standard will give the user a more precise understanding of the characteristics and limitations of the equipment. Thus. Electromedical diagnostic and patient monitoring equipment). Suite 220. Technical Advisory sub-Group for IEC/SC 62D/WG1 (now IEC/SC 62D/MT22. Comments and suggested revisions should be sent to AAMI. The working group also decided to include some requirements and informative text which apply only to the AAMI standard and to refer to ANSI/AAMI EC57:1998(R)2003 as a new normative reference. Copying. Glebe Road. Medical electrical equipment – Part 2-47: Particular requirements for the safety. if met.S. Arlington. EC57’s more comprehensive set of reporting requirements are used instead of those found in the IEC standard. working as the U. These include areas which the IEC standard does not address. As previously noted. and test specifications which. To remain relevant. This standard is written primarily for manufacturers of ambulatory monitoring systems to give them clear labeling. performance.S. a worldwide organization for standardization. the AAMI ECG/Ambulatory Electrocardiograph Working Group. Medical electrical equipment – Part 2 – 47: Particular requirements for the safety. IEC 60601-2-47:2001 was developed by IEC Subcommittee (SC) 62D. During the course of revision. such understanding is conducive to a safer and more efficacious use of the device. the AAMI ECG/Ambulatory Electrocardiograph Working Group decided to adopt IEC 60601-2-47. including essential performance. In addition. performance. but it does provide important information about the development and intended use of the document. The objective of this standard is to provide minimum labeling. Ambulatory electrocardiographs. consensus on the international standard and otherwise participated in the drafting of that document. networking. This standard reflects the conscientious efforts of concerned health care professionals and medical device manufacturers to develop a standard for those performance levels that can be reasonably achieved at this time. design.Background of ANSI/AAMI adoption of IEC 60601-2-47:2001 This standard was developed by the ECG/Ambulatory Electrocardiograph Working Group of the AAMI Electrocardiograph (ECG) Committee. and distribution prohibited. including essential performance. NOTE—This foreword does not contain provisions of the American National Standard. should promote patient safety and diagnostic quality of the results. The concepts incorporated in this standard should not be considered inflexible or static. This is a revision of ANSI/AAMI EC38:1998. VA 22201-4795. which is administered by AAMI on behalf of the International Electrotechnical Commission (IEC). of ambulatory electrocardiographic systems. of ambulatory electrocardiographic systems (ANSI/AAMI EC38:2007). like any other. which is based in large part on the AAMI standard noted above. This standard. must be reviewed and updated periodically to assimilate progressive technological developments. . Suggestions for improving this standard are invited. it must be modified as technological advances are made and as new data comes to light. 1110 N. Electromedical Equipment. and safety requirements that will help ensure a reasonable level of clinical efficacy and patient safety in the use of ambulatory electrocardiographs (ECGs). AAMI standards served as the basis of the international standard.

the MIT database. The accuracy of ventricular ectopic beat (VEB) detection shall be tested using the AHA database.AAMI Deviations to IEC 60601-2-47:2001 The U. its ability to do so shall be tested using the CU database. 50. 8 min each. The CU database contains digitized single-channel ECG recordings with rhythm changes labeled.2. 30 min each).101. ECG – AHA database: The American Heart Association Database for Evaluation of Ventricular Arrhythmia Detectors (80 records. the MIT database.1 Standard databases to be used to evaluate automated analyses IEC Requirement Four standard databases are available from three sources for evaluating ambulatory analyzers and have been widely used for that purpose. – NST database: The Noise Stress Test Database (12 records. deviations to IEC 60601-2-47:2001 that appear in ANSI/AAMI EC38:2007 are: SECTION EIGHT – ACCURACY OF OPERATING DATA AND PROTECTION AGAINST HAZARDOUS OUTPUT *50. supplied with the MIT database). and distribution prohibited.1 Standard databases to be used for the performance test The accuracy of QRS detection shall be tested using the AHA database. networking. supplied with the MIT database).2. If the EQUIPMENT is claimed to detect ventricular flutter or fibrillation (VF). 30 min each. The first three databases consist of digitized excerpts of two-channel Holter recordings. and the NST database.101. – CU database: The Creighton University Sustained Ventricular Arrhythmia Database (35 records. the AHA database and the MIT database. with each beat labeled.101. its ability to do so shall be tested using the MIT database and the NST database. – MIT database: The Massachusetts Institute of Technology–Beth Israel Hospital Arrhythmia Database (48 records. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.101. Copying.2 Record-by-record results For each record. xi . 30 min each). Replaced by ANSI/AAMI Requirement See ANSI/AAMI EC57:1998/(R)2003.2 Performance reporting requirements IEC Requirement 50. If the EQUIPMENT is claimed to detect supraventricular ectopic beats (SVEBs) or atrial flutter or fibrillation (AF).S. *50. the following statistics shall be reported: − QRS sensitivity and positive predictivity. This is a complete list of those databases that are both adequate as a basis for evaluation and generally available at present. and the NST database.

Symbols and abbreviations used in these tables are: G gross statistic (calculated for the entire database) +P positive predictivity R required (if the reported) Se sensitivity VEB ventricular ectopic beat EQUIPMENT is claimed to detect the abnormality. gross statistics summarizing the performance of the EQUIPMENT under test on each of the databases used for testing shall be reported in tabular format as shown in table 102 and 103. and distribution prohibited. . the statistic shall be SVEB supraventricular ectopic beat TP true positive TN true negative FP false positive FN false negative Table 102 – Reporting requirements for standard analyzer outputs Statistic xii AHA DB MIT DB NST DB CU DB QRS Se (G) R R R – QRS +P (G) R R R – VEB Se (G) R R R – VEB +P (G) R R R – VEB Couplets Se (G) R R – R VEB Couplets +P (G) R R – R VEB Short Runs Se (G) R R – R VEB Short Runs +P (G) R R – R VEB Long Runs Se (G) R R – R VEB Long Runs +P (G) R R – R © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. For each record. 50.101. any of the following statistics that are relevant to the capabilities shall also be reported: − VF episode sensitivity and positive predictivity.2.− VEB sensitivity and positive predictivity. networking. Copying. − AF episode sensitivity and positive predictivity.101. − SVEB sensitivity and positive predictivity.3 EQUIPMENT ’s claimed Statistics Based on the record-by-record reports required by 50.2.2.

If the evaluation of the equipment under test is performed using signals converted into analogue form and supplied to the normal analogue inputs of the equipment.4 c). the equipment's automatic gain control (AGC) will be allowed to adjust the gain automatically.Table 103 – Reporting requirements for optional analyzer outputs Statistic AHA DB MIT DB NST DB CU DB VF Episode Se (G) R R – R VF Episode +P (G) R R – R SVEB Se (G) – R R – SVEB +P (G) – R R – 50.2. i. and (where applicable) SVEB Se and SVEB +P. − reformatting (i. without rewinding manner in which the evaluator no way is to be construed as a performs its analysis. If the evaluation is performed using digital data and the AGC is not digital but part of the analogue front end of the equipment.e.2. VEB Se.e. shall be used to derive QRS Se.101.e. the equipment may simulate its AGC capabilities by an alternative method.) Use of the standard databases for test Each record shall be supplied end (that is. conversion to a sampling rate different from that used in the standard database files). − filtering performed by software or hardware not employed in the normal operating mode of the equipment under test. Pre-processing includes. − rescaling (altering the signal amplitude.4 Test methods (NOTE a) For maximum legibility the text in subclauses a) to d) is not italicized. This requirement applies only to the presents ECG samples to the EQUIPMENT under test and in restriction on the manner in which the EQUIPMENT normally If the digitized ECG signals from the database records are pre-processed in any way before presenting them as input to the equipment under test. VEB +P. to the EQUIPMENT continuously from the beginning to the or ‘fast forwarding’). xiii . Copying. but is not limited to: − resampling (i. − conversion from digital to analogue signals.101.101. and distribution prohibited. The evaluator shall then run the 'xform' (or equivalent) program to adjust the ECG ’s gain based on the instructions provided by the program.2. The protocol described in 50.. changing the gain). This process shall be repeated until ‘no gain change’ is announced and the equipment under test shall then automatically proceed with the ECG analysis. Beat-by-beat comparisons. following the protocol described in 50.. where applicable. This alternative method allows the ‘test mode’ that generates the ‘test annotations’ to emit an announcement that a ‘gain adjustment’ would be required prior to proceeding with analyzing the ECG for each patient record. sample precision or numeric coding). networking. conversion of byte order.4 d) shall be used to derive VF and AF episode Se and +P. the pre-processing shall be disclosed in sufficient detail to permit a third party to reproduce the test. QRS +P. This announcement should instruct the evaluator to adjust the gain of the ECG for either one or both of the ECG channels. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.

101. or for any other reason. − S: supraventricular ectopic beat (SVEB): an atrial or nodal (functional) premature or escape beat or an aberrant atrial premature beat. . They represent the absence of a beat label. beat labels are defined as follows: − N: any beat that does not fall into the S. − Q: paced beat. equipment may also generate U labels. networking.101. Copying. To perform the comparison between the test annotation files and the reference annotation files.2. In order to mark segments during which an equipment's analysis is suspended (shut down) because of excessive noise or signal loss. or Q categories described below (normal or bundle branch block beat). − F: fusion of a ventricular and a normal beat. Within annotation files. − O: a pseudo-beat label generated at any other time. ‘rxr’. The reference annotation files distributed with the databases and used as input to these programs may not be altered in any way. In the AHA database.2. fusion of a paced and normal beat or a beat that cannot be classified.4 c): − U: label that marks a segment of unreadable data. to preserve a one-toone correspondence between beat labels. the xiv © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.2. and to end 150 ms before the following beat label.4 d) require that. except that (where applicable) corrected reference annotation files obtained from the database suppliers may be substituted for those originally distributed with the databases.b) Use of annotation files The test protocols described in 50. for each record. Beat labels are never paired with U labels during beat-by-beat comparisons. A manufacturer using programs other than the ones released by MIT shall show the equivalence of his programs.102. − V: ventricular ectopic beat (VEB): a ventricular premature beat. which is assumed for testing purposes to begin 150 ms after the previous beat label. the output of the equipment has been recorded in an annotation file (the test annotation file) in the same format as the reference annotation file for that record. ‘epic’. a pair of U labels marks the beginning and end of each unreadable segment. Extra beats are sometimes detected (false positive ‘QRS’) and true beats are sometimes missed (false negative QRS). If either the reference or the test annotation file contains an extra beat label that has no match in the other file. the programs supplied on the MIT-BIH Arrhythmia Database CD-ROM (programs ‘bxb’. and distribution prohibited. In the MIT database. pseudo-beat labels are added to those in the reference and test annotation files. a single U label marks the (approximate) centre of each unreadable segment.4 c) through 50. ‘mxm’) or later versions released by MIT or programs with equivalent function and output shall be used. Any automated procedure for doing so is acceptable as long as it is disclosed. V. In order to perform beat-by-beat comparisons. In beat-by-beat comparisons. all beat labels are paired up. The equipment need not produce this file directly. There are two types: − X: a pseudo-beat label generated during a segment marked as unreadable. U labels appear in the database where beats cannot be located because of excessive noise or signal loss in both channels. F. an “R-on-T” ventricular premature beat or a ventricular escape beat. Other labels are needed to facilitate the beat-by-beat comparison process defined in 50.

appropriate O or X label is paired with the extra label. Those which mark segments of atrial flutter or fibrillation (AF.4 c)). it is possible that a matching test beat label may be placed before the beginning of the test period. All such beat label pairs are counted. others are ignored. If T is within 150 ms of the beginning of the test period. Rhythm labels mark segments of ventricular flutter or fibrillation (VF) in the AHA and MIT databases: − [ : beginning of VF − ] : end of VF Beat labeling is discontinued between ‘[’ and ‘]’ labels. or for VF and AF comparisons (50. the test annotation at t is not counted (t is set to the time of the next test beat label before going on to step 2). see the documentation which accompanies each database) are used for evaluation of AF detection. c) Beat-by-beat comparison During a beat-by-beat comparison.101. O and X labels are not used in runby-run comparisons (50. Additional rhythm labels mark changes in rhythm in the MIT database. VF segments are excluded from beat-by-beat comparisons. The end product of a beat-by-beat comparison is a matrix in which each element is a correct count of the number of beat label pairs of the appropriate type: Table 104 – Beat-by-beat matrix Algorithm label Reference label n s v f q o x N Nn Ns Nv Nf Nq No Nx S Sn Ss Sv Sf Sq So Sx V Vn Vs Vv Vf Vq Vo Vx F Fn Fs Fv Ff Fq Fo Fx Q Qn Qs Qv Qf Qq Qo Qx O On Os Ov Of Oq X Xn Xs Xv Xf Xq The beat-by-beat comparison is performed according to the following method: 1) Set the variable T to the time of the first reference beat label after the end of the learning period and set the variable t to the time of the first test beat label after the end of the learning period.4 d)). © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. This corresponds to a QRS detection error – either a false detection (if the extra label is in the test annotation file) or a missed beat (if it is in the reference annotation file). since it is not necessary in these instances to pair individual beat labels. xv . Beat labels are never paired with rhythm labels.101. it is counted as a match (t is set to the time of the matching test beat label before going on to step 2). if t is within 150 ms of the beginning of the test period and there is no matching reference beat label after the beginning of the test period. the candidate beat is considered to have been missed or to be an extra detection. reference beat labels and equipment beat labels are matched up in pairs. the absolute value of the difference between the equipment's estimate of the time of occurrence of a beat and the time as recorded in the reference annotation file shall not exceed 150 ms. and distribution prohibited. including those that involve O or X labels.2. networking. On the other hand. If matching does not occur within this window. Copying. If this occurs. To be considered a match.2. Set all elements of the matrix to zero.

c) The matrix element corresponding to the beat label pair which was generated in step b) is incremented. The results of beat-by-beat comparisons are used to derive QRS sensitivity and positive predictivity: QTP = Nn + Ns + Nv + Nf + Nq + QFN = No + Nx Sn + Ss + Sv + Sf + Sq + So + Sx + Vn + Vs + Vv + Vf + Vq + Vo + Vx -+ Fn + Fs + Fv + Ff + Fq + Fo + Fx + Qn + Qs + Qv + Qf + Qq Qo + Qx QFP = On + Os + Ov + Of + Oq + Xn + Xs + Xv + Xf + Xq QRS Se = QTP/(QTP + QFN) xvi QRS +P = QTP/(QTP + QFP) © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. There are now two possibilities: (i) If T is closer to t than to t′. The variable t is reset to the value of t′. During unreadable segments. The extra label is paired with an O or X ‘pseudo-beat’ label. The extra reference beat label is paired with an O or X "pseudo-beat" label. set T′ to the time of the next reference beat label (or to a time beyond the end of the record if there are no more reference beat labels). The variable T is reset to the time of the next reference beat label. Test beat labels generated during true VF segments are not counted for these purposes. at all other times. this possibility shall be taken into account by software designed to perform beat-by-beat comparisons. In principle. . and distribution prohibited. the beat labels at T and t are paired. Reference beat labels present during equipmentmarked VF segments are paired with O pseudo-beat labels and counted like all other missed beats. There are again two possibilities: (i) If t is closer to T than to T′. pseudo-beat labels are O. Copying. networking. set t′ to the time of the next test beat label (or to a time beyond the end of the record if there are no more test beat labels). the procedure has to keep track of segments that have been marked as unreadable or as VF in either the reference or the test annotation file. During the derivation of the matrix. (ii) Otherwise. b) If t does not precede T. an unreadable segment or a VF segment may begin during the learning period.2) One of the following cases shall apply: a) If t precedes T. d) Steps b) and c) are repeated until both t and T are set to times beyond the end of the record. the equipment has missed the beat at T. and t is within 150 ms (the match window) of T. the beat labels at T and t are paired. The variable T is reset to the value of T ′. (ii) Otherwise. and t is within 150 ms of T. The variable t is reset to the time of the next test beat label. pseudo-beat labels are X. the test beat label at t is an extra detection.

This test requires the production of an annotation file based on the equipment's outputs. any other reference episodes are counted as false negatives. any other algorithm-marked episodes are counted as false positives. The leading edge overshoot shall be less than 10 %. VF and AF episode sensitivity and positive predictivity are derived in the usual way. false negatives and false positives derived according to the above-mentioned methods. a VF comparison shall be performed. each algorithm-marked episode for which overlap exists is counted as a true positive for purposes of determining VF episode positive predictivity.5.3 mV/s. From the counts of true positives. thus a detector is neither penalized nor rewarded for its treatment of ventricular fusion beats and ambiguous beats. networking. For equipment which is claimed to detect AF. an AF comparison shall be performed.67 Hz to 40 Hz shall be between 140 % and 70 % (+3 dB to –3 dB) of the response at 5 Hz. xvii . Each reference episode for which overlap exists is counted as a true positive for purposes of determining VF episode sensitivity. so that a detector's treatment of ambiguous beats does not influence its measured SVEB detection performance. containing (at a minimum) the times when the equipment has determined that episodes of VF have begun or ended. note that Qs is excluded from SVTP and SVFP. and distribution prohibited.1 mV referred to the baseline before the pulse. Segments labeled as atrial flutter in the reference annotation files shall be excluded from this comparison. SVEB sensitivity and positive predictivity are similarly defined: SVTP = Ss SVFN = Sn + Sv + Sf + Sq + So + Sx SVFP = Ns + Vs + Fs + Os + Xs SVEB Se = SVTP/(SVTP + SVFN) SVEB +P = SVTP/(SVTP + SVFP) Again. and either b) The amplitude response to sinusoidal signals within the frequency range 0. Replaced by ANSI/AAMI Requirement See ANSI/AAMI EC57:1998/(R)2003. Copying. Similarly. Overlap exists during any interval in which both the reference and algorithm annotations indicate that VF is in progress. *51.The results of beat-by-beat comparisons are used to derive VEB sensitivity and positive predictivity: VTP = Vv VFN = Vn + Vs + Vf + Vq + Vo + Vx VFP = Nv + Sv + Ov + Xv VEB Se = VTP/(VTP + VFN) VEB + P = VTP/(VTP + VFP) Note that VTP and VFP do not include Fv or Qv. This test is performed in the same manner as the VF comparison with the substitution of ‘AF’ for each occurrence of ‘VF’ in the description above. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.9 Frequency response IEC Requirement The EQUIPMENT shall meet the following requirements: a) Response of a RECORDER to a 3 mV 100 ms rectangular pulse shall not show a baseline amplitude displacement after the pulse of more than 0. d) VF and AF comparisons For equipment which is claimed to detect VF. The slope outside the pulse shall be less than 0.

the lowest peak-to base amplitude of the 1. If the manufacturer claims ST segment measurement capability for the EQUIPMENT . networking. Record during 2 min while slowly changing the repetition rate from 60/min to 70/min in order to ensure that the full range of amplitude variability resulting from sample points missing the peak of the pulse will be obtained. 200 ms base width pulses with a repetition rate of 1/s.5 mV × 200 ms triangular wave. b) The p-v amplitude responses at the frequencies of 0. 10 Hz. replace the lower test frequency of 0. If the manufacturer claims ST segment measurement capability for the EQUIPMENT . lower cutoff frequency shall be 0. and distribution prohibited. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg.5 mV. or c) Responses to all pulses of a 1. 20 Hz. replace the upper test frequency of 40 Hz by 55 Hz. shall be within 60 % to 110 % of the response to a train of 1. 2 Hz. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg.05 Hz or its functional equivalent. Compliance is checked by the following tests: The system input is at the record. 20 Hz and 40 Hz. and then a single 3 mV 100 ms rectangular pulse.5 mV. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. replace in the above series of tests the upper test frequency of 40 Hz by 55 Hz.1 mV from the baseline preceding the pulse.5 mV. xviii © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 1 Hz.67 Hz.67 Hz by 0. Then adjust the base width of the triangular pulses to 40 ms.67 Hz by 0. replace in the above series of tests the lower test frequency of 0.5 mV. Continue recording for at least another 20 s of zero volt baseline. .05 Hz for a first-order high-pass filter or its functional equivalent. Verify the following measures on the hard copy record: a) The output baseline following the 3 mV rectangular pulse is displaced no more than 0. the output is measured on the system's hard copy ECG a) Record at least 20 s of zero volt baseline. 2 Hz.5 mV. Copying. which simulates a series of narrow R-waves.5 mV 200 ms base width triangle pulse train. record for at least 5 s a 2 mV p-v sinusoidal signal at 0. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. 5 Hz.05 Hz or its functional equivalent. c) The lowest peak-to base amplitude of the 1. 40 ms triangular pulse train. PATIENT ELECTRODE . 40 ms base width triangle pulse train is no less than 60 % of the highest peak-to base amplitude of the 1. the response to the 1.If the manufacturer claims ST segment measurement capability for the EQUIPMENT . the upper cut-off frequency shall be at least 55 Hz.5 mV × 40 ms triangular wave shall be within 80 % to 110 % of the 1.3 mV/s. b) With the test set-up of figure 104. The slope outside the region of the pulse does not exceed 0. c) Record for at least 5 s a train of triangular 1. Repeat this for 1 Hz. 40 ms base width triangle pulse train is no less than 80 % of the highest peak-to-base amplitude of the 1.67 Hz.5 mV 200 ms base width triangle pulse train. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. 40 Hz are between 70 % and 140 % of the response at 5 Hz. 10 Hz. 200 ms triangular pulses.

they shall be disclosed. The slope everywhere outside the rectangular pulse shall be less than 0.1 mV from the baseline preceding the pulse. replace in the above series of tests the upper test frequency of 40 Hz by 55 Hz. 10 Hz. the output is measured on the system's hard copy ECG a) Record at least 20 s of zero volt baseline. 5 Hz. and then a single 3 mV 100 ms rectangular pulse.) Amplitude displacement of the baseline shall not be greater than 0.5 mV  4 0 ms triangular wave shall be within 80% to 110% of the 1. 40 ms triangular pulse train.0 to -3. The input of the system is at the PATIENT ELECTRODES .5 mV.5 mV  2 00 ms triangular wave. 2 Hz. 20 Hz. a) Response to a 3 mV. 200 ms base width pulses with a repetition rate of 1/s. Record during 2 min while slowly changing the repetition rate from 60/min to 70/min in order to ensure that the full range of amplitude variability resulting from sample points missing the peak of the pulse will be obtained. and distribution prohibited.67 Hz to 40 Hz shall be between 112% and 71% (+1. Repeat this for 1 Hz. 10 Hz. which simulates a series of narrow R-waves.) 2. 20 Hz and 40 Hz. networking. And c) Responses to all pulses of a 1. The slope outside the region of the pulse does not exceed 0. 1. If the manufacturer claims that the EQUIPMENT is capable of recording ECGs from infants weighing less than 10 kg.5 mV. 2 Hz. shall be within 60% to 110% of the response to a train of 1. the response to the 1. b) With the test set-up of figure 104. Then adjust the base width of the triangular pulses to 40 ms. 40 Hz are between 70 % and 140 % of the response at 5 Hz. xix .0 mV/s in devices not claiming ST measurement capabilities. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.67 Hz.0 dB) of the response at 5 Hz with a tolerance of ±5%. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. 1 Hz. If special filtering compensation circuits are used in the recorder and/or the reproducer. 200 ms triangular pulses. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. b) The p-v amplitude responses at the frequencies of 0. Compliance is checked by the following tests: The system input is at the record. Continue recording for at least another 20 s of zero volt baseline.1 mV from onset to end in devices claiming ST measurement capabilities or 1.Replaced by ANSI/AAMI Requirement The overall system frequency response shall meet the specifications described. The leading edge overshoot shall be less than 10%. 100 ms rectangular pulse shall meet the following criteria. b) Amplitude response to sinusoidal signals within the frequency range of 0. PATIENT ELECTRODE . record for at least 5 s a 2 mV p-v sinusoidal signal at 0.30 mV/s in devices claiming ST measurement capabilities or 2. replace the upper test frequency of 40 Hz by 55 Hz. Verify the following measures on the hard copy record: a) The output baseline following the 3 mV rectangular pulse is displaced no more than 0.0 mV in devices not claiming ST measurement capabilities. the output of the system is measured on the system’s hard copy ECG record. Copying.5 mV. c) Record for at least 5 s a train of triangular 1.67 Hz.3 mV/s.) 3.

5 mV 200 ms base width triangle pulse train. a third pulse having an amplitude of 20 mV and a duration of 0. 40 ms base width triangle pulse train is no less than 80 % of the highest peak-to-base amplitude of the 1. xx © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. to P2.2 mV from the height of the sine wave peak immediately preceding the pulse. Record at least 30 s with the sinusoidal generator settings of item b) above and a repetition rate of the pulses of 100/min and verify that for every pulse. the device shall suitably record ECG signals in the presence of pacemaker pulses having amplitudes between 2 mV and 250 mV. e) Confirm on playback that for all pulses the height of the second peak of the sine wave after the pulse does not differ more than 0.2) mV p-v sinusoidal signal is present across the 111 Ω resistor.0 ± 0. Replaced by ANSI/AAMI Requirement The device labeling shall clearly indicate whether or not the device is intended for recording ECG signals in the presence of implanted pacemaker pulses.1 ms. Compliance is checked by the following test: a) Connect the EQUIPMENT to the circuit of figure 106.c) The lowest peak-to base amplitude of the 1. . with the positive PATIENT ELECTRODE connection for each channel connected to P1 and the negative ELECTRODE connection for each channel. durations between 0. durations between 0.5 mV 200 ms base width triangle pulse train.11 Function in the presence of pacemaker pulses IEC Requirement If the manufacturer claims that the AMBULATORY RECORDER is capable of recording ECG signal in the presence of implanted pacemaker pulses. and a frequency of up to 300 pulses per min. Compliance for such EQUIPMENT is checked by the following test: Tests with four different pulses with a rise time of <100 μ s shall be made: a first pulse having an amplitude of 2 mV and a duration of 2. the lowest peak-to base amplitude of the 1.0 ms. *51. as well as the reference ELECTRODE connection.1 ms and a fourth pulse having an amplitude of 2 mV and a duration of 0.5. a mark at least 2 mm high is printed on the hard copy record at the same repetition frequency and same inter-pulse interval as the pulses input into the EQUIPMENT .0 ms. and distribution prohibited.5 mV. b) Adjust the sine wave generator so that a 10 Hz (2.1 ms and 2. the EQUIPMENT shall produce a visible recording for pacemaker pulses with amplitudes between 2 mV and 200 mV.0 ms and a rise time of <100 μs. networking.1 ms and 2. a rise time ≤ 100 μ s and a repetition rate of 100 pulses/min.5 mV. d) Reverse the positive and negative ELECTRODE connections in a) and repeat the recording. with rise times equal to 10% of the pulse width but not greater than 100 μs. Copying.0 ms ± 0.1 ms. 40 ms base width triangle pulse train is no less than 60 % of the highest peak-to base amplitude of the 1. If the manufacturer claims that EQUIPMENT is capable of recording the activity of an implanted pacemaker. the function of the EQUIPMENT shall not be adversely affected by the operation of an implanted pacemaker. If the device is indicated as intended for such recording. The pulse generator adds 200 mV ± 25 mV pulses with a duration of 1. a second pulse having an amplitude of 200 mV and a duration of 2.0 ms. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. c) Record at least 30 s.

durations between 0.01 ms. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. with rise times equal to 10% of the pulse width but not greater than 100 μs. l) PATIENT ELECTRODES of step (j) and Confirm on playback that for all pacemaker pulse waveforms. PATIENT ELECTRODES of step (e) and g) Adjust the rectangular pulse generator so that a 2. a) Connect the EQUIPMENT to the circuit of figure 106. h) Record at least 30 s connections in (g) and repeat the recording.1 ms.1 ± 0. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor.01 ms. Compliance for such EQUIPMENT is checked by the following test: Testing shall be performed using four different pacemaker pulse waveforms: the first pulse having an amplitude of 250 mV and a duration of 2. to P2. a second pulse having an amplitude of 250 mV and a duration of 0.2) mV p-v sinusoidal signal is present across the 111 Ω resistor.0 mV and a duration of 0. e) Reduce the pulse width to 0. and a frequency of up to 300 pulses per min. c) Record at least 30 s d) Reverse the positive and negative ELECTRODE connections in (a) and repeat the recording.0 mV and a duration of 0. i) Reverse the positive and negative j) Reduce the pulse width to 0. reverse the positive and negative record at least 30 s of the test waveform. with the positive PATIENT ELECTRODE connection for each channel connected to P1 and the negative ELECTRODE connection for each channel. The device shall produce an indication of the pacemaker pulse that is visible on the hard copy.0 ms and a fourth pulse having an amplitude of 2.1 ms and 2. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor. the device shall suitably record pacemaker activity for pacemaker pulses having amplitudes between 2 mV and 250 mV.0 ± 0. a second pulse having an amplitude of 250 mV and a duration of 0.1 ms.0 mV ± 0.0 mV and a duration of 2.0 mV and a duration of 2.Compliance is checked by the following test: Testing shall be performed using four different pacemaker pulse waveforms: the first pulse having an amplitude of 250 mV and a duration of 2. f) For each channel. as well as the reference ELECTRODE connection.0 ms and a fourth pulse having an amplitude of 2. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor. networking. xxi . Adjust the rectangular pulse generator so that a 250 mV ± 25 mV pulses with a duration of 2. Copying. Adjust the rectangular pulse generator so that a 250 mV ± 25 mV pulses with a duration of 2. b) Adjust the sine wave generator so that a 10 Hz (2. This indication shall have a height of at least 2mm. If the manufacturer indicates that the device is suitable for use to record pacemaker activity.2 mV from the height of the sine wave peak immediately preceding the pulse. with the positive PATIENT ELECTRODE connection for each channel connected to P1 and the negative ELECTRODE connection for each channel. a) Connect the EQUIPMENT to the circuit of figure 106. and record at least 30 s of this test waveform.0 ms.1 ms.0 ms ± 0. and distribution prohibited.2 mV pulses with a duration of 2. and record at least 30 s of this test waveform. reverse the positive and negative record at least 30 s of the test waveform. a third pulse having an amplitude of 2. the height of the second peak of the sine wave after the pulse does not differ more than 0.0 ± 0.1 ms.0 ms.0 ms ± 0.0 ms ± 0. b) Adjust the sine wave generator so that a 10 Hz (2.1 ± 0. as well as the reference ELECTRODE connection. ELECTRODE k) For each channel.1 ms. to P2. a third pulse having an amplitude of 2.1 ms.2) mV p-v sinusoidal signal is present across the 111 Ω resistor.1 ms.0 ms.

and record at least 30 s of this test waveform. Additional ANSI/AAMI Requirements (not in IEC standard) See: 51. e) Reduce the pulse width to 0. including essential performance.5.16 Special considerations for high speed superimposition display (optional) on page 17.5.17 Lead Definitions on page 18. Medical electrical equipment – Part 2-27: Particular requirements for the safety. of electrocardiac monitoring equipment xxii © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. ELECTRODE k) For each channel.0 mV ± 0. ANSI/AAMI EC57:1998/(R)2003.0 ms ± 0.5. and distribution prohibited.5. Copying.2 mV pulses with a duration of 2.20 Full disclosure (miniature displays) on page 19. networking. 51.1 ± 0. reverse the positive and negative record at least 30 s of the test waveform.01 ms.5. Medical electrical equipment – Part 2-27: Particular requirements for the safety of electrocardiac monitoring equipment Replaced by ANSI/AAMI Reference IEC 60601-2-27:2005. i) Reverse the positive and negative j) Reduce the pulse width to 0. 51. and record at least 30 s of this test waveform. 51.19 Patient event marks on page 19. Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms IEC Reference IEC 60601-2-27:1993. PATIENT ELECTRODES of step (e) and g) Adjust the rectangular pulse generator so that a 2. . Appendix L (normative) References – Publications mentioned in this standard ANSI/AAMI Addition ASSOCIATION FOR THE ADVANCEMENT OF MEDICAL INSTRUMENTATION.01 ms. a mark at least 2 mm high is printed on the hard copy record at the same repetition frequency and same inter-pulse interval as the pulses input into the EQUIPMENT .18 Baseline stability on page 19. f) For each channel. l) PATIENT ELECTRODE s of step (j) and Confirm on playback that for every pulse. reverse the positive and negative record at least 30 s of the test waveform. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor. h) Record at least 30 s connections in (g) and repeat the recording.c) Record at least 30 s d) Reverse the positive and negative ELECTRODE connections in (a) and repeat the recording. 51.1 ± 0.1 ms.

thereby ‘improving’ its apparent performance. supplied with the MIT database). supplied with the MIT database). The rationale for the related ANSI/AAMI requirements. A MBULATORY ECG SYSTEMS are often designed to provide optimal performance when used with a skilled technician who interacts with the program. 30 min each). With human intervention allowed.101. It is acceptable to perform the tests of the analysis software on such a testbed if the port of the analysis software to the EQUIPMENT is validated separately. that is. The evaluation methodology requires the combination of the EQUIPMENT with its interface. thereby permitting verification of test results when the same test data are used. MIT database: The Massachusetts Institute of Technology–Beth Israel Hospital Arrhythmia Database (48 records. In principle. Full disclosure of the procedure for generating annotation files enables an independent (thirdparty) evaluator to use the procedure. evaluations of these EQUIPMENT s have to be performed without human intervention. deviations. Thus. 8 min each. and distribution prohibited. CU database: The Creighton University Sustained Ventricular Arrhythmia Database (35 records. For this reason. AMBULATORY ECG SYSTEMS AHA database: The American Heart Association Database for Evaluation of Ventricular Arrhythmia Detectors (80 records. IEC Rationale 50 Accuracy of operating data 50. for this reason. Specifications for complete evaluation techniques for AMBULATORY ECG scanning systems are outside the scope of this Standard. perfect results are achievable in principle for any EQUIPMENT which provides ‘full-disclosure’ output. including (where appropriate) an explanation for the U.1 Use of standard databases Four standard databases are available from three sources for evaluating and have been widely used for that purpose. evaluations that allow human intervention measure only the persistence and expertise of the OPERATOR and are of no value in assessing the performance of the EQUIPMENT . can be found in Annex AA of this document. It also permits the use of additional test data of the evaluator’s choice as such data become available. It is common practice to present the digitized ECG s from the database directly to the analysis software because this allows accurate and reproducible test results. xxiii . 30 min each. such evaluations are neither required nor encouraged.S. networking. 30 min each) NST database: The Noise Stress Test Database (12 records. 50. Full disclosure will provide a disincentive for having the interface do anything other than straightforward translations of the EQUIPMENT s normal outputs into standard annotation files. Copying. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.Annex AA (informative) Guidance and rationale Replaced rationale The following rationale from the IEC standard for the requirements sections listed above is provided for reference purposes. a strictly reproducible ‘hands-off’ evaluation is required. the interface might include significant analytical components when processing the outputs of the EQUIPMENT .101 Automated analysis A credible evaluation has to be reproducible.

MIT Room 20A113. and CU databases). On the other hand.2 Performance reporting requirements. Plymouth Meeting.The first three databases consist of digitized excerpts of two-channel Holter recordings. . Since performance is highly dependent on the characteristics of the particular ECG s that are analyzed. The CU database contains digitized single-channel ECG recordings with rhythm changes labeled. For this reason. ANSI/AAMI Rationale See 50. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.5 h of unannotated signals per record immediately preceding the 30 min test periods) is used. An evaluation of VF detection using the 80 records of the AHA DB and the 48 records of the MIT database should supplement the required CU database evaluation. 56100 Pisa. however. ESC. Via Trieste. The exclusion of records with paced beats is permitted only for EQUIPMENT s that are not designed to analyze paced analogue ECG recordings made without pacer artifact detection or enhancement. Computer Laboratory. Well-defined QRS complexes necessary for a beat-by-beat comparison are not present during VF segments. Italy. This standard allows the manufacturer to try each database PATIENT with all three of the standard gains on both channels of the ECG to see which gain setting produces the ‘best’ results. 50. since the CU database does not contain a sufficient number of signals that are likely to provoke false VF detections. 41. Cambridge.2 Performance reporting requirements Several issues cannot be addressed adequately using existing testing methodology. PA 19462. only the final 35 min of each record (equivalent to the standard version) may be presented to the EQUIPMENT under test.101.101. MA 02139. evaluations have to be performed using standard recordings in order for the results of those evaluations to have value for purposes of comparison among EQUIPMENT s or against a performance standard. and distribution prohibited. The standard allows for the manufacturer to change the gain settings manually.1 page 26.101. the ESC DB is available from the CNR Institute of Clinical Physiology. networking. The ANSI/AAMI document EC38:1998 provides further guidance about evaluation of analysis software (particularly for runs of ectopic beats). Most EQUIPMENT s need a certain amount of time to learn the underlying rhythm. If the long version of the AHA database (containing 2. MIT –BIH Database Distribution. which are marked by rhythm labels in the reference annotation files and excluded from beat-by-beat comparisons. USA (AHA DB). with each beat labeled. NST. these segments are included in tests of consecutive VEB and SVEB detection and VF detection. xxiv USE of standard databases and 50. These databases are available through: ECRI. Copying. This is a complete list of those databases that are both adequate as a basis for evaluation and generally available at present. The incidence and variety of VF in the AHA and MIT databases are insufficient to allow those databases to serve as substitutes for the CU database for the purposes of evaluating the VF detection capability of the EQUIPMENT . on the grounds that the original analogue tapes do not reproduce pacemaker artifacts with sufficient fidelity to permit use of common techniques for recognition of these artifacts in ‘live’ signals. a 5 min learning period is allocated at the beginning of each record and is excluded from calculated performance statistics. This approximates the automatic gain control feature of most AMBULATORY ECG SYSTEMS in use today. 77 Massachusetts Avenue. The digital databases reproduce the analogue tapes with sufficient fidelity to permit use of techniques used for recognition of pacemaker artifacts by EQUIPMENT s designed to analyze analogue ambulatory ECG tapes. Outside North America. USA (MIT. 5200 Butler Pike. however.

© 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.05 Hz high pass filter has been considered acceptable. et al. p. the phase response of a one pole 0. which is most critical for low frequency response. Historically. The impulse response test is also employed to simulate the R-wave and to observe readily whether the monitor produces baseline changes following the impulse that. Second. the persistent problem of highfrequency noise from power line frequencies and from muscle artifact can be reduced with a 40 Hz bandwidth. 1 SIMONSON. This standard proposes test methods to evaluate both the frequency response and the ability of the AMBULATORY ECG EQUIPMENT to deal with ECG -like signals such as a triangular wave simulating the R-wave. ANSI/AAMI Rationale See 51. The baseline offset allowed following the impulse represents the drop that can be expected from a 0. Differentiation between normal and abnormal in electrocardiography.3 mm across a typical ST interval of 100 ms and would not be clinically significant.30 mV/s translates at standard gain and speed to a change of 0.05 Hz filter. and distribution prohibited. E. St. used these data to justify the 0.3 Hz.67 Hz low frequency bound in the 1990 AHA recommendations. might result in artifactual displacements of the ST segment and lead to false interpretations regarding the presence of ischemia..3 mV × s). Selecting +3 dB as the upper limit for frequency response accuracy allow for this boost.407.9 Incorrect output Frequency response Complete specification of frequency response should address phase distortion. the primary purposes of the ambulatory ECG (which are [a] to identify rhythms and [b] to reveal displacements of the ST segment necessary to identify ischemic episodes) can be adequately accomplished without a higher frequency response. 1961. networking.5 51.67 Hz) exists for 99 % of adults 90 % of the time. Variability of the electrocardiogram in normal young men. 38. The slope requirement of 0.V.5. 0. at 40 Hz an accurate measure would require a time base of 400 mm/s. SIMONSON. Tests at higher frequencies are not proposed because measuring phase shift for frequencies above 25 Hz would be difficult at best at the required time base of 25 mm/s. The low-frequency response of 0. 1949. p. vol. These studies indicate that 44 bpm encompasses more than 99 % of adult heart rates with intra-individual RR interval variation less than 126 ms.67 Hz is based upon heart rate data from the Simonson studies 1 . Bailey et al.IEC Rationale 51 Protection against hazardous output 51. particularly as a typical QRS complex has an impulse value closer to 0. The impulse response requirement tests this capability with a relatively easy-to-apply procedure. xxv .158.9 Frequency response. Copying.5. First. E. Am Heart. The high-frequency response limit of 40 Hz is based on two considerations. There is a boost in the frequency response of analogue tape RECORDER s at approximately 0. Thus a lower bound of 40 bpm (0. with a real ECG input signal.1 mV × s (cf. Louis:C. page 30. Allowing a 40 % reduction in the peak value of the triangular input signal corresponds to the expected reduction due to a digital sampling system that stores one 24 h channel in about 10 million samples.2 Hz to 0. Mosby Co.

19 Patient event marks on page 33.5.5. Copying. . and distribution prohibited. 51.17 Lead definitions on page 32. References – Informative Additional ANSI/AAMI Reference Section (not in IEC standard) See References – Informative on page 34. xxvi © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.5. 51.5.5. 51.Additional ANSI/AAMI Rationale (not in IEC standard) See: 51.20 Full disclosure (miniature displays) on page 33. networking. 51.16 Special considerations for high speed superimposition display (optional) on page 31.18 Baseline stability on page 32.

International Standard IEC 60601-2-47 has been prepared by subcommittee 62D: Electromedical equipment. Any divergence between the IEC Standard and the corresponding national or regional standard shall be clearly indicated in the latter. technical reports or guides and they are accepted by the National Committees in that sense. 3) The documents produced have the form of recommendations for international use and are published in the form of standards. The text of this standard is based upon the following documents: FDIS Report on voting 62D/408/FDIS 62D/411/RVD Full information on the voting for the approval of this standard can be found in the report on voting indicated in the above table. and distribution prohibited. To this end and in addition to other activities. Copying. the IEC publishes International Standards. networking. The IEC collaborates closely with the International Organization for Standardization (ISO) in accordance with conditions determined by agreement between the two organizations. The object of the IEC is to promote international co-operation on all questions concerning standardization in the electrical and electronic fields. This publication has been drafted in accordance with the ISO/IEC Directives. any IEC National Committee interested in the subject dealt with may participate in this preparatory work. xxvii . as nearly as possible. 4) In order to promote international unification. Part 3. governmental and non-governmental organizations liaising with the IEC also participate in this preparation. Their preparation is entrusted to technical committees.INTERNATIONAL ELECTROTECHNICAL COMMISSION ___________ MEDICAL ELECTRICAL EQUIPMENT – Part 2-47: Particular requirements for the safety. of ambulatory electrocardiographic systems FOREWORD 1) The IEC (International Electrotechnical Commission) is a world-wide organization for standardization comprising all national electrotechnical committees (IEC National Committees). International. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. Annex AA is for information only. 5) The IEC provides no marking procedure to indicate its approval and cannot be responsible for any equipment declared to be in conformity with one of its standards. technical specifications. 2) The formal decisions or agreements of the IEC on technical matters express. 6) Attention is drawn to the possibility that some of the elements of this International Standard may be the subject of patent rights. The IEC shall not be held responsible for identifying any or all such patent rights. IEC National Committees undertake to apply IEC International Standards transparently to the maximum extent possible in their national and regional standards. of IEC technical committee 62: Electrical equipment in medical practice. an international consensus of opinion on the relevant subjects since each technical committee has representation from all interested National Committees. including essential performance.

replaced by a revised edition. the publication will be • • • • reconfirmed. withdrawn. 2 OF THE GENERAL STANDARD OR THIS PARTICULAR STANDARD : SMALL The committee has decided that the contents of this publication will remain unchanged until 2005. A bilingual edition of this publication may be issued at a later date. explanations. Copying. . or amended. general statements. introductions. and distribution prohibited. At this date. compliance with which can be tested. networking. xxviii © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.In this Particular Standard the following print types are used: – requirements. advice. – TERMS DEFINED IN CLAUSE CAPITALS . and definitions: in roman type. – notes. – test specifications: in italic type. exceptions and references: in smaller type .

INTRODUCTION This Particular Standard concerns the safety of AMBULATORY ELECTROCARDIOGRAPHIC SYSTEMS . this annex does not form part of the requirements of this Standard. A “General guidance and rationale” for the requirements of this Particular Standard is included in annex AA. as amended by its amendment 1 (1991) and its amendment 2 (1995). and distribution prohibited. Copying. It amends and supplements IEC 60601-1 (second edition 1988): Medical electrical equipment – Part 1: General requirements for safety. xxix . It is considered that knowledge of the reasons for these requirements will not only facilitate the proper application of the standard but will. expedite any revision necessitated by changes in clinical practice or as a result of developments in technology. hereinafter referred to as the General Standard. An asterisk (*) by a clause or subclause number indicates that some explanatory notes are given in annex AA of this Particular Standard. However. in due course. networking. The requirements of this Particular Standard take priority over those of the General Standard. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.

Single user license only. Copying. networking. . and distribution prohibited.

ECG 1.3 Particular Standards Addition: This Particular Standard refers to IEC 60601-1 (1988): Medical electrical equipment – Part 1: General requirements for safety. Copying. Within the scope of this standard are systems of the following types: a) systems that provide continuous recording and continuous analysis of the ECG allowing full re-analysis giving essentially similar results.101.2 Object Replacement: The object of this Particular Standard is to establish particular requirements for the safety. and distribution prohibited. 1. as amended by its amendment 1 (1991) and its amendment 2 (1995). Medical electrical equipment covered by IEC 60601-2-25 and IEC 60601-2-27 are excluded from the scope of this standard. The type of storage media used is irrelevant with regard to this standard. 1 . of AMBULATORY ELECTROCARDIOGRAPHIC SYSTEMS . If the ambulatory electrocardiographic system offers automatic ECG analysis. minimal performance requirements for measurement and analysis functions apply. including essential performance. b) systems that provide continuous analysis and only partial or limited recording not allowing a full re-analysis of the ECG . networking. safety requirements for AMBULATORY ELECTROCARDIOGRAPHIC SYSTEMS .American National Standard ANSI/AAMI EC38:2007 MEDICAL ELECTRICAL EQUIPMENT – Part 2-47: Particular requirements for the safety. of ambulatory electrocardiographic systems SECTION ONE – GENERAL The clauses and subclauses of this section of the General Standard apply except as follows: 1 Scope and object This clause of the General Standard applies except as follows: 1. This standard does not apply to systems that do not continuously record and analyze the (for example. or record and analyze the ECG simultaneously. The systems may first record and store the ECG and analyze it later on a separate unit. including essential performance.1 Scope Addition: This Particular Standard specifies the particular as defined in 2. ‘intermittent event recorders’). © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. The safety aspects of this standard apply to all types of systems falling in one of the abovementioned categories.

‘Amendment’ means that the clause or subclause of the General Standard is amended as indicated by the text of this Particular Standard. tables and figures which are additional to those of the General Standard are numbered starting from 101. The numbering of sections and subclauses of this Particular Standard corresponds to that of the General Standard. networking. Copying. and additional items aa). Medical electrical equipment – Part 1: General requirements for safety – 2. . where it is intended that any part of the General Standard. bb). subclauses. BB. 2 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. although possibly relevant. applies without modification. additional appendices are lettered AA. and distribution prohibited. although possibly not relevant. IEC 60601-1 is referred to in this Particular Standard either as the General Standard including its collateral standards or as the General Requirement(s).5 Collateral Standards Addition: IEC 60601-1-2:1993. both of which may contain an analysis function NOTE—This EQUIPMENT is often referred to as Holter monitoring equipment after its inventor Dr.For brevity. Changes to the text of the General Standard are specified by the following ‘Replacement’ means that the clause or subclause of the General Standard is replaced completely by the text of this Particular Standard. is not to be applied. a statement to that effect is given in this Particular Standard. ‘Addition’ means that the text of this Particular Standard is additional to the requirements of the General Standard. Clauses. Norman Holter. clause or subclause of the General Standard. the section. Where there is no corresponding section.103 PLAYBACK EQUIPMENT EQUIPMENT for monitoring and documenting functions into which data from the RECORDER is fed NOTE—This EQUIPMENT is usually stationary and commonly includes computing facilities. 1. Collateral standard: Electromagnetic compatibility – Requirements and tests *2 Terminology and definitions This clause of the General standard applies except as follows: Additional definitions: 2.. etc. 2. etc.101 AMBULATORY ELECTROCARDIOGRAPHIC SYSTEM ( EQUIPMENT ) AMBULATORY RECORDER and a PLAYBACK EQUIPMENT .102 AMBULATORY RECORDER recording EQUIPMENT worn or carried by the PATIENT including associated cables for recording or recording and analyzing heart action potentials ELECTRODES and 2. The term ‘this Standard’ is used to make reference to the General Standard and this Particular Standard taken together. clause or subclause in this Particular Standard.

networking.104 ELECTROCARDIOGRAM ( ECG ) visual record of heart action potentials [IEC 60601-2-25:1993. Examples: [IEC 60601-2-25:1993.104] 2. 2.109 LEAD WIRE ( S ) cable connected between the ELECTRODE and the AMBULATORY RECORDER .105 LEAD ELECTRODE and LEAD WIRE combination(s) used for a certain recording of Einthoven limb LEAD II . Identification.2. 3 .101] 2.109] 2. Copying. not forming part of any ELECTROCARDIOGRAPH LEAD [IEC 60601-2-25:1993. definition 2. modified] 2. Classification This clause of the General Standard applies except as follows: 5. definition 2. definition 2.108 PATIENT CABLE multiwire cable and associated connector(s) to connect the ELECTRODES to the AMBULATORY RECORDER [IEC 60601-2-25:1993.110 CONTINUOUS RECORDER EQUIPMENT which performs 5 continuous analysis and/or recording of the ECG .107] 2. definition 2. and distribution prohibited.6 Amendment: Delete all but 6 CONTINUOUS OPERATION . marking and documents This clause of the General Standard applies except as follows: © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.106 PATIENT ELECTRODE means in contact with a specified part of the body to detect heart action voltage in combination with another means [IEC 60601-2-25:1993.107 NEUTRAL ELECTRODE reference point for differential amplifiers and/or interference suppression circuits.103. Unipolar chest LEAD V5 ECG . definition 2.

cc) Clear instructions shall be provided for any use of the 4 RECORDER in wet environments. b NOTE—The desired polarity assignments are presented here. and so constructed or marked as to avoid incorrect connection to the EQUIPMENT .2 Instructions for use Additional items: aa) Advice shall be given on the following: 1) the procedures necessary for safe operation. 4) the possible hazard caused by the summation of LEAKAGE CURRENTS when several EQUIPMENTS are interconnected by coupling and/or a multiple portable socket-outlet . and distribution prohibited. . should not contact other conductive parts including earth. 2) the type of electrical installation to which the EQUIPMENT may be safely connected. bb) Clear instructions shall be provided if a specific type of battery or battery charging procedure has to be used in order to fulfill the requirements of this Particular Standard. Also.1 Marking on the outside of EQUIPMENT or EQUIPMENT parts Additional item: aa) LEAD identification The LEAD ( S ) shall be permanently marked in such a manner that the proper LEAD can be directly determined at both the ELECTRODE attachment end. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. networking. but equipment can deviate as long as the deviation is properly labeled. 6. drawing attention in the case of TYPE B APPLIED PARTS to the safety hazard which may occur as a result of an inadequate electrical installation. the LEAD ( S ) shall be color coded according to one of the color coding schemes of table 101. b Code 2 is widely used in North America – see AHA guidelines of 1985. 3) that conductive parts of ELECTRODES and associated connectors for TYPE BF APPLIED PARTS or TYPE CF APPLIED PARTS .8. Table 101 – lead color codes ELECTRODE Channel 1 Positive ELECTRODE Negative Channel 2 Positive ELECTRODE Negative Channel 3 Positive ELECTRODE ELECTRODE ELECTRODE Negative ELECTRODE NEUTRAL ELECTRODE Code 1 a Code 2 green red red white white brown yellow black orange orange blue blue black green a Code 1 is widely used in Europe and internationally.6. the channel assignment shall be clearly annotated on the EQUIPMENT for reference. Copying. including the NEUTRAL ELECTRODE . including the connection to any POTENTIAL EQUALIZATION CONDUCTOR . If independent bipolar leads are being used.

or whether the report presents this information episode by episode. whether ranges of heart rates. whether there are OPERATOR selectable detection criteria for ST segment shifts (such as displacement and slope parameters). ee) The manufacturer shall disclose the method for calculating the heart rate. Copying.g. how frequently ST segment shifts are summarized in the report (e. 20. ranges of displacements and/or slope values during each episode are reported. and distribution prohibited. the manufacturer shall disclose in the operating manual or physician's guide the following: • • • • whether the ST analysis is performed on all LEADS using any or all calibration signals. b) A relative humidity of 10 % to 95 %.1 Environment Amendment: For RECORDERS : a) An ambient temperature range of 10 °C to 45 °C. ff) The manufacturer shall disclose the method for determining a pause.2.. SECTION TWO – ENVIRONMENTAL CONDITIONS The clauses and subclauses of this section of the General Standard apply except as follows: *10 Environmental conditions 10. 5 .dd) The EQUIPMENT labeling shall clearly indicate whether or not its use is intended for infants weighing less than 10 kg. types of episodes (elevation or depression). without condensation. SECTION THREE – PROTECTION AGAINST ELECTRIC SHOCK HAZARDS The clauses and subclauses of this section of the General Standard apply except as follows: 20 Dielectric strength This clause of the General Standard applies except as follows: 20. gg) If the equipment is designed to detect and/or measure ST segment shifts. hourly) and whether numbers of episodes.3 Values of test voltages Addition: © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. networking. and durations of episodes are reported.2 Particular requirements for EQUIPMENT with an APPLIED PART Amendment: B-b Does not apply to EQUIPMENT .

CLASS II . 6 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. If the recorder is normally used with a pouch. SECTION FOUR – PROTECTION AGAINST MECHANICAL HAZARDS The clauses and subclauses of this section of the General Standard apply except as follows: 21 Mechanical strength *21. the RECORDER shall not be damaged after being subjected to shocks resulting from an 0. hardwood >600 kg/m 3 ) lying flat on a rigid base such as a concrete floor and making solid contact with the base on every face.500 V ( CLASS I . The RECORDER shall be unaffected and shall resume normal data acquisition within 60 s of the shock.8 m onto a 50 mm thick hardwood board (for example.B-d1 The test voltage shall be 1. 21. The RECORDER shall not suffer obvious damage as a result of this test and shall meet the requirements of this Particular Standard. If the recorder is normally used with a pouch. and INTERNALLY POWERED EQUIPMENT ) between F . Compliance is tested as follows: The RECORDER is dropped once from a height of 75 mm onto a 50 mm thick hardwood board (for example. networking. as above).5 Replacement: Data acquisition by the AMBULATORY RECORDER may be interrupted during shock but data acquired prior to the shock shall be unaffected and normal data acquisition shall resume within 60 s after the completion of the following test.8 m drop onto a hard surface on any face. . the same type of pouch can be used during testing. hardwood >600 kg/m 3 ) which lies flat on a rigid base such as a concrete floor. the same type of pouch can be used during the testing. or when not operating. This test does not apply if the I/O parts cannot be connected to external EQUIPMENT while the device is PATIENT connected.TYPE APPLIED PARTS and SIGNAL INPUT PARTS and SIGNAL OUTPUT PARTS . The requirement from the General Standard for a minimum reference voltage of U = 250 V for INTERNALLY POWERED EQUIPMENT does not apply. B-d2 The test voltage between F .6 Replacement: During transport or storage. Compliance is tested as follows: The RECORDER is allowed to fall freely once from each of three different starting positions from a height of 0. edge and corner.TYPE APPLIED PARTS and ENCLOSURE other than SIGNAL INPUT PARTS shall be determined by the MAINS VOLTAGE of the device and table V of the General Standard. and distribution prohibited. Copying. edge or corner (pouch may be used.

201 Emissions 36. terminated in a load EQUIPMENT during the test (see Immunity Addition to paragraph 4: Examples of indications of possible SAFETY HAZARDS include changes in operating state.1 Replacement: RECORDER shall comply with the requirements of CISPR 11.202.1.SECTION FIVE – PROTECTION AGAINST HAZARDS FROM UNWANTED OR EXCESSIVE RADIATION The clauses and subclauses of this section of the General Standard apply except as follows: *36 ELECTROMAGNETIC COMPATIBILITY IEC 60601-1-2:1993 applies except as follows: 36. 7 .202.201. Additionally.1 Requirements Amendment: © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. a level of ±8 kV shall apply for non-conductive ACCESSIBLE PARTS .202. *36. networking. 36. Group 1. *36.201. A level of ±6 kV shall apply for contact discharge to conductive ACCESSIBLE PARTS and coupling planes. 36.1 Electrostatic discharge Replacement: Equipment and/or systems shall comply with IEC 61000-4-2. The EQUIPMENT shall return to the previous operating mode within 10 s without loss of any stored data.1.201.2). Class B.7 Replacement : PATIENT . Copying. irrecoverable loss or change of stored data. and distribution prohibited.COUPLED EQUIPMENT AND / OR SYSTEM shall be tested with transducers.202.2.202 the PATIENT CABLES .2 Radiated radio-frequency electromagnetic fields 36. LEAD ( S ) and ELECTRODES attached to the EQUIPMENT and simulating the PATIENT (figures 101 and 102). Signal input/output cables (if applicable) shall be attached to the item a) of 36.1 Radio frequency (RF) emissions 36.2.

6 Magnetic fields Addition: The EQUIPMENT shall be exposed to a magnetic field intensity of 3 A/m magnetic flux density of 1 gauss at three times the line frequency. b) For EQUIPMENT . and distribution prohibited. . The performance requirements of this Particular Standard shall be met and no data loss shall occur.f. c) Delete.000 MHz. networking. the test frequency shall be swept from 80 MHz to 1.202. SECTION SEVEN – PROTECTION AGAINST EXCESSIVE TEMPERATURES AND OTHER SAFETY HAZARDS The clauses and subclauses of this section of the General Standard apply. Copying. field level shall be Test conditions Amendment: a) 80 % amplitude modulation at a single modulation frequency in the range of 1 Hz to 5 Hz shall be used. AMBULATORY RECORDER 36.a) The 3 V/m.2. This addition refers to IEC 61000-4-8. The EQUIPMENT cable shall be bundled non-inductively to 1 m overall length and the signal cable (if applicable) and POWER SUPPLY CORD (if applicable) shall be arranged horizontally and vertically from the EQUIPMENT according to figure 102. SECTION SIX – PROTECTION AGAINST HAZARDS OF IGNITION OF FLAMMABLE ANAESTHETIC MIXTURES The clauses and subclauses of this section of the General Standard apply. SECTION EIGHT – ACCURACY OF OPERATING DATA AND PROTECTION AGAINST HAZARDOUS OUTPUT The clauses and subclauses of this section of the General Standard apply except as follows: 50 Accuracy of operating data Addition: 8 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 36. The r.2 shall comply with IEC 61000-4-3.202. not applicable.

5 Pauses The total number of pauses detected based upon an OPERATOR selectable absolute threshold or manufacturer selected parameter shall be reported.101. and runs of three or more VEBs. include the following labeling: – time of strip.3. 50.3. single SVEBs. paired SVEBs. 50. multichannel ECG strips shall be available with each report in sufficient quantity to support all meaningful clinical conclusions.101.101. paired VEBs.101. The number of minutes (and optionally seconds) that were analyzed on each channel shall be reported (the manufacturer may substitute the amount of time not analyzed).101.3. – heart rate on strip. The location and duration of the longest of pauses shall be reported. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.3. ECG strips shall.7 ECG hard copy O PERATOR selectable.1 Heart rate Lowest. *50. 50. 50.6 ST segment shifts If the manufacturer claims that EQUIPMENT is capable of detecting and measuring ST segment shifts.1 Standard databases to be used to evaluate automated analyses See ANSI/AAMI EC57:1998/(R)2003. rate and duration (either beat totals or time duration) of each episode shall be reported.2 Supraventricular ectopy Totals for SVEBs. The report shall summarize each item at least once for each hour of the ambulatory procedure and then as a procedure total at the end of the procedure. and the duration of runs (either number of beats or time duration) shall be reported. – strip annotation.101. 50.3 Physician report – minimum requirements Any abnormality in the items listed below that an EQUIPMENT system is capable of detecting shall be reported. runs of SVT and some form of SVT duration (either beat totals or time duration) shall be reported.101 Automated analysis EQUIPMENT having automated ECG analysis is required to fulfill the requirements of this section. The report shall summarize each item of the ambulatory procedure at regular time intervals defined by the manufacturer and then as a procedure total at the end of the procedure. *50.*50. The summary information shall also reflect the total number of heart beats detected.101. Bradycardia episodes (heart rate less than 50/min for 15 s or manufacturerselected parameters or user-defined parameters) shall be reported. single VEBs. and distribution prohibited. 9 .4 Bradycardia data Hourly presentation of the total of bradycardia episodes is required. minimally.3.3 Ventricular ectopy Totals of ventricular ectopic beats (VEBs). and highest heart rates shall be reported.101. 50. LEAD configuration for each channel shall be provided either with each ECG strip or as part of the procedure settings information. 25 mm/s. networking. 50. For episodes of ventricular tachycardia.3. *50. Copying.101. The report shall also list all OPERATOR selected parameters.101. Summary information shall include the total number of each event that occurred during the procedure. a suitable report with the manufacturer-claimed parameters shall be generated and included in the ACCOMPANYING DOCUMENTS .2 Performance reporting requirements See ANSI/AAMI EC57:1998/(R)2003. mean. specifying rate and duration of the episodes.3.

c.Additionally. g) Join the negative PATIENT ELECTRODE connection of each channel through P2 to the reference LEAD WIRE through a parallel combination of a 51 k Ω resistor and a 47 nF capacitor. and distribution prohibited. wait for 30 s and repeat the recording. Requirement of differential voltage for digital recorders shall be 10 mV p-v.) offset voltage of ±300 mV. *51 Protection against hazardous output *51. The indicated time-varying output signal amplitude referred to input shall not change by more than 10 % or 50 μV.4 mV. d) Set switch S3 to position B and use switch S4 to subtract a 300 mV offset. each ‘page’ of ECG strips shall contain the PATIENT identification. The indicated time-varying output signal amplitude referred to input shall not change by more than 10 % or 50 μV. whichever is greater. Each channel calibration signal shall be present in each report for which a subsequent ST segment analysis is to be done. S3 in position A and the positive PATIENT ELECTRODE connection for each channel joined to P1. 10 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. Record the triangular wave. wait for 30 s and repeat the recording.1 Input dynamic range EQUIPMENT shall be capable of responding to and displaying differential voltages of 6 mV peakto-valley (p-v) in amplitude for analog recorders (when set to the 5 mm/mV gain setting) and varying at a rate of 125 mV/s in the presence of a direct current (d. c) Set switch S3 to position B and use switch S4 to add an offset voltage of 300 mV. Requirement of differential voltage for analog recorders shall be 6 mV p-v.4 Hz triangular wave with an amplitude of 10 mV p-v (figure 103) into the test circuit of figure 104 with switches Sl and S2 closed. S3 in position A and the positive PATIENT ELECTRODE connection for each channel joined to P1. Compliance is checked by the following test: a) Set the gain to 5 mm/mV. A ‘page’ in this context might be a single ECG strip from an ECG strip recorder or several strips contained on a A4-sized or ‘letter’-size sheet of paper. Feed a 10. Feed a 10.5.c. Compliance is checked by the following test: f) Set the gain to 5 mm/mV. e) Confirm that in the playback signal display the triangular waves have a minimum amplitude pv of 5.) offset voltage of ±300 mV. whichever is greater.4 Hz triangular wave with an amplitude of 6 mV p-v (figure 103) into the test circuit of figure 104 with switches Sl and S2 closed. networking. Copying. b) Join the negative PATIENT ELECTRODE connection of each channel through P2 to the reference LEAD WIRE through a parallel combination of a 51 k Ω resistor and a 47 nF capacitor. . Record the triangular wave. EQUIPMENT shall be capable of responding to and displaying differential voltages of 10 mV peakto-valley (p-v) in amplitude for digital recorders (when set to the 5 mm/mV gain setting) and varying at a rate of 125 mV/s in the presence of a direct current (d.5 Incorrect output Replacement: *51.

Within this outer shield the placements of the fixture. *51. f) Repeat the test with offset voltages of 300 mV and –300 mV respectively. Compliance is checked by the following test: a) Refer to the test circuit of figure 104. Set the interference signal EQUIPMENT © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. open S5.0 mV. The common mode rejection capability is defined as the ratio of the p-v value of the interfering SUPPLY MAINS frequency to the p-v value of the resulting signal in any ECG input channel.5.5. c) Connect the PATIENT ELECTRODE connections of the first channel to P1 and P2.3 Common mode rejection Common mode rejection shall be at least 60 dB for a sinusoidal signal at the SUPPLY MAINS frequency and at least 45 dB at twice the SUPPLY MAINS frequency. 11 . This requirement shall be met across the total required d. offset range capabilities. *51. i) Set switch S3 to position B and use switch S4 to subtract a 300 mV offset. An additional earth referenced shield shall enclose the entire test set-up. networking. d. a) Use the manufacturer’s recommended PATIENT CABLE or equivalent. the PATIENT CABLE and the foil wrapping of the PATIENT CABLE must all be well controlled and repeatable after fixture is calibrated in order to minimize changes in the fixture’s calibration. d) Measure the output amplitude on the manufacturer’s PLAYBACK EQUIPMENT . wait for 30 s and repeat the recording. put S3 in position A. Compliance is checked by the following test: Refer to the test circuit of figure 105.2 Input impedance The input impedance shall be greater than 10 MΩ for the frequency specified in the test and for all input channels. Alternatively: Test with a 4 Hz sine wave.c. Connect all other PATIENT ELECTRODE connections to P6. b) Close switches S1 and S2. Copying. either continuous or consisting of isolated cycles repeated once a second. and distribution prohibited. g) Repeat all these tests for all other ECG channels. The same driven shield shall enclose the various resistor/capacitor networks. referred to input. e) Open S1 and measure the output amplitude change. The steady-state output amplitude shall not decrease by more than 6 %. The PATIENT CABLE shall be enclosed throughout its entire length by a similar foil shield connected to the shield driven by the simulated SUPPLY MAINS frequency source. Apply a 10 Hz sinusoidal signal of 5 mV amplitude p-v across P1 and P2. except where the PATIENT CABLE enters and shall conform to the contours of the EQUIPMENT within 3 mm. j) Confirm that in the playback signal display the triangular waves have a minimum amplitude pv of 9. offset source and switches.c. wait for 30 s and repeat the recording. The foil shall fully enclose the EQUIPMENT . Enclose the under test in a conductive foil wrap and connect this to earth.h) Set switch S3 to position B and use switch S4 to add an offset voltage of 300 mV.

4 Gain accuracy The output at all possible gain settings shall be reproduced with a maximum amplitude error of ±10 % compared to the test signal. and close the outer shield. *51. b) Connect all PATIENT ELECTRODE LEADS to a common node. Open the outer shield. the foil around the PATIENT CABLE . Compliance is checked by the following test: Apply a 5 Hz. if supplied.422 V p-v. each one in series with a parallel combination of a 51 k Ω resistor. referred to the input. 10 min. Copying. d) Repeat the test with switches S1 to Sn closed. At twice the SUPPLY MAINS frequency the measured output shall not exceed 4 mV p-v with a generator voltage of 1. SUPPLY MAINS The measured output during each test period shall not exceed 4 mV p-v at SUPPLY MAINS frequency with a generator voltage of 4V p-v. Any supply frequency notch filters in the EQUIPMENT shall be disabled during these tests. and the fixture are all in place inside the earth grounded outer shield in the positions they will occupy when the EQUIPMENT is added after calibration. offset of 300 mV and –300 mV in series with the imbalance impedance by opening Sa and testing with Sb in each position. Switches S1 to Sn inclusive are open. The resulting test voltage across C t will then be 1/2 of the signal generator value. even if it requires software not available to the customer to do so.initially at the SUPPLY MAINS frequency. .c. Adjust C t until C t + C x = 100 pF. 2 min. Compliance is checked by the following test: Apply a 5 Hz. 12 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.05 mV at the ECG inputs shall be available for analog systems for the purpose of calibrating the system. The outer shield must be closed in the position that will be used during the actual tests. and the PATIENT CABLE . Repeat the tests at twice the frequency. This adjustment must be done while the EQUIPMENT under test is totally removed from the test set-up. taking into account any possible aliasing and the reproducer’s maximum OPERATOR selectable playback speed. *51.5. The signal may be in the form of a step or a pulse with a rise time of <5 ms (time from 10 % to 90 %).5 Gain stability One minute after energizing the EQUIPMENT . Repeat with offset in series with each input. the gain change shall not exceed 3 % over a 24 h period (in stable ambient conditions). Record sufficient signal to allow measurement of worst case interference.5. a 47 nF capacitor and a switch. networking.0 ± 0. and distribution prohibited. switch Sa is closed. 30 min. 5 min. 45 min and 60 min) and once every following hour up to 24 h. Connect any common or reference ELECTRODE . connect the EQUIPMENT . Connect the low side of the generator to earth. 2 mV p-v sinusoidal signal to all ECG input channels. c) Repeat the test with a d. *51.5. 2 mV p-v sinusoidal signal to all ECG input channels for a period of 24 h. Apply the interference test signal to the common node through a 100 pF capacitor. through a 51 k Ω resistor in parallel with a 47 nF capacitor to the same common node. The output shall comply with the above requirement at each possible gain setting. in turn. 20 min. With each possible gain setting. Compliance is checked by inspection.6 Amplitude calibration A calibration signal equivalent to a voltage of 1. verify that the output is within the requirements at any time during the first hour (or test at 1 min.

5.2 mV p-v referred to input.0 mV/s in devices not claiming ST measurement capabilities. record for 2 min.5. Join only one positive PATIENT ELECTRODE connection to P1 at a time. Ignore the first 10 s and last 10 s of the recording. 100 ms rectangular pulse shall meet the following criteria. At the highest gain possible. switch S3 in position A and the positive PATIENT ELECTRODE connections for each channel joined to P1.5.*51.1 mV from onset to end in devices claiming ST measurement capabilities or 1. *51. 1) 2) 3) Amplitude displacement of the baseline shall not be greater than 0. b) Join the reference PATIENT ELECTRODE connections for each channel through P2 to the reference LEAD WIRE through a 51 k Ω resistor in parallel with a 47 nF capacitor. The output of the channels with the positive shall not exceed 0. the output of the system is measured on the system’s hard copy ECG record. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.9 Frequency response The overall system frequency response shall meet the specifications described. a) Response to a 3 mV. S5 open. The leading edge overshoot shall be less than 10%.2 mV p-v. e) Repeat this for as many channels as can be recorded. The input of the system is at the PATIENT ELECTRODES . d) Reconnect all but one of the positive PATIENT ELECTRODE connections from P1 to P2.0 mV in devices not claiming ST measurement capabilities. c) Adjust the signal generator to produce a sinusoidal signal of an amplitude of 4 mV p-v and a frequency of 10 Hz across P1 and P2. The p-v noise level shall be within the limit for at least nine of the ten intervals. The slope everywhere outside the rectangular pulse shall be less than 0.7 System noise The internal noise referred to input shall not exceed 50 μV p-v over any 10 s period when all inputs are connected through a 51 kΩ resistor in parallel with a 47 nF capacitor in series with each PATIENT ELECTRODE connection.30 mV/s in devices claiming ST measurement capabilities or 2. then check the output for noise levels in each interval. EQUIPMENT shall not produce in any channel an output Compliance is checked by the following test: a) Connect the RECORDER to the test circuit of figure 104 with switches S1 and S2 closed. Record at least 10 s of signal. networking. and then connect all PATIENT ELECTRODE connections including the right leg connection together. they shall be disclosed. 13 . Copying. Record at least 10 s of signal. Do not connect the input signal generator and the 100 pF capacitor for this test. shall be operating at the appropriate SUPPLY MAINS frequency during this test. as shown in figure 105. and distribution prohibited. PATIENT ELECTRODE connection connected to P2 *51. Compliance is checked by the following test: Insert in series which each PATIENT ELECTRODE connection a 51 k Ω resistor in parallel with a 47 nF capacitor. Any SUPPLY MAINS frequency notch filters in the EQUIPMENT .8 Multichannel crosstalk The crosstalk between the channels of the referred to input greater than 0. If special filtering compensation circuits are used in the recorder and/or the reproducer. if so equipped. Divide the remaining 100 s into 10 intervals of 10 s each.

67 Hz. replace in the above series of tests the upper test frequency of 40 Hz by 55 Hz. record for at least 5 s a 2 mV p-v sinusoidal signal at 0. which simulates a series of narrow R-waves. 10 Hz. 10 Hz.b) Amplitude response to sinusoidal signals within the frequency range of 0. Verify the following measures on the hard copy record: a) The output baseline following the 3 mV rectangular pulse is displaced no more than 0. the lowest peak-to base amplitude of the 1.67 Hz.5 mV. 2 Hz. b) The p-v amplitude responses at the frequencies of 0. 40 Hz are between 70 % and 140 % of the response at 5 Hz. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. And c) Responses to all pulses of a 1. 40 ms triangular pulse train. shall be within 60% to 110% of the response to a train of 1. Continue recording for at least another 20 s of zero volt baseline. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. 2 Hz. . If the manufacturer claims that the EQUIPMENT is capable of recording ECGs from infants weighing less than 10 kg. If the manufacturer claims that the EQUIPMENT is capable of recording ECG s from infants weighing less than 10 kg. 50 μV p-v sinusoidal signal shall yield a visual recorded deflection at a time base of 25 mm/s and a gain setting of 10 mm/mV. 200 ms base width pulses with a repetition rate of 1/s. Compliance is checked by measurement.5 mV.0 dB) of the response at 5 Hz with a tolerance of ±5%. Record during 2 min while slowly changing the repetition rate from 60/min to 70/min in order to ensure that the full range of amplitude variability resulting from sample points missing the peak of the pulse will be obtained. 20 Hz and 40 Hz.5 mV 200 ms base width triangle pulse train. c) The lowest peak-to base amplitude of the 1. b) With the test set-up of figure 104. c) Record for at least 5 s a train of triangular 1. 40 ms base width triangle pulse train is no less than 60 % of the highest peak-to base amplitude of the 1. The slope outside the region of the pulse does not exceed 0. Compliance is checked by the following tests: The system input is at the record. 1 Hz.5 mV. and then a single 3 mV 100 ms rectangular pulse. Then adjust the base width of the triangular pulses to 40 ms. the response to the 1. and distribution prohibited. 14 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.3 mV/s.5.5 mV.1 mV from the baseline preceding the pulse. 20 Hz. replace the upper test frequency of 40 Hz by 55 Hz.5 mV. 40 ms base width triangle pulse train is no less than 80 % of the highest peak-to-base amplitude of the 1. Repeat this for 1 Hz. *51. 5 Hz.67 Hz to 40 Hz shall be between 112% and 71% (+1.0 to -3. networking.5 mV  2 00 ms triangular wave.5 mV  4 0 ms triangular wave shall be within 80% to 110% of the 1. 200 ms triangular pulses. the output is measured on the system's hard copy ECG a) Record at least 20 s of zero volt baseline. Copying.10 Minimum feature size A 10 Hz.5 mV 200 ms base width triangle pulse train. PATIENT ELECTRODE .

reverse the positive and negative record at least 30 s of the test waveform. and distribution prohibited.0 ms.2 mV pulses with a duration of 2. and record at least 30 s of this test waveform.1 ms. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor. Compliance for such EQUIPMENT is checked by the following test: Testing shall be performed using four different pacemaker pulse waveforms: the first pulse having an amplitude of 250 mV and a duration of 2. a) Connect the EQUIPMENT to the circuit of figure 106. ELECTRODE k) For each channel. networking.1 ms and 2.2) mV p-v sinusoidal signal is present across the 111 Ω resistor.0 ms ± 0. reverse the positive and negative record at least 30 s of the test waveform. and record at least 30 s of this test waveform. If the device is indicated as intended for such recording.0 ms. a second pulse having an amplitude of 250 mV and a duration of 0.1 ms. the device shall suitably record pacemaker activity for pacemaker pulses having amplitudes between 2 mV and 250 mV. a third pulse having an amplitude of 2. i) Reverse the positive and negative j) Reduce the pulse width to 0. c) Record at least 30 s d) Reverse the positive and negative ELECTRODE connections in (a) and repeat the recording.0 mV and a duration of 0. the device shall suitably record ECG signals in the presence of pacemaker pulses having amplitudes between 2 mV and 250 mV. f) For each channel.1 ± 0.0 ms. as well as the reference ELECTRODE connection. b) Adjust the sine wave generator so that a 10 Hz (2. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor.0 mV and a duration of 0. l) PATIENT ELECTRODES of step (j) and Confirm on playback that for all pacemaker pulse waveforms.0 mV and a duration of 2. Compliance is checked by the following test: Testing shall be performed using four different pacemaker pulse waveforms: the first pulse having an amplitude of 250 mV and a duration of 2. with rise times equal to 10% of the pulse width but not greater than 100 μs.5.1 ms. PATIENT ELECTRODES of step (e) and g) Adjust the rectangular pulse generator so that a 2. If the manufacturer indicates that the device is suitable for use to record pacemaker activity. a second pulse having an amplitude of 250 mV and a duration of 0.0 ms and a fourth pulse having an amplitude of 2. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. a third pulse having an amplitude of 2.1 ms. Copying.0 ms. This indication shall have a height of at least 2mm. durations between 0. the height of the second peak of the sine wave after the pulse does not differ more than 0.0 ± 0.01 ms.1 ms and 2. durations between 0. with the positive PATIENT ELECTRODE connection for each channel connected to P1 and the negative ELECTRODE connection for each channel. with rise times equal to 10% of the pulse width but not greater than 100 μs. e) Reduce the pulse width to 0.1 ms.0 ms ± 0.0 ms and a fourth pulse having an amplitude of 2. Adjust the rectangular pulse generator so that a 250 mV ± 25 mV pulses with a duration of 2. and a frequency of up to 300 pulses per min.01 ms. and a frequency of up to 300 pulses per min. h) Record at least 30 s connections in (g) and repeat the recording.1 ms. The device shall produce an indication of the pacemaker pulse that is visible on the hard copy. to P2.0 mV ± 0.2 mV from the height of the sine wave peak immediately preceding the pulse. 15 .*51.11 Function in the presence of pacemaker pulses The device labeling shall clearly indicate whether or not the device is intended for recording ECG signals in the presence of implanted pacemaker pulses.1 ± 0.0 mV and a duration of 2.

and distribution prohibited. c) Record at least 30 s d) Reverse the positive and negative ELECTRODE connections in (a) and repeat the recording. i) Reverse the positive and negative j) Reduce the pulse width to 0.2. and record at least 30 s of this test waveform.a) Connect the EQUIPMENT to the circuit of figure 106. Adjust the rectangular pulse generator so that a 250 mV ± 25 mV pulses with a duration of 2.01 ms.25 mm/s and 5 mm/mV (for example. Compliance is checked by measurement according to the following test: Using an optical magnifier with a resolution of at least 0.0 mV ± 0.01 ms. If the scale factors are less than or equal to 6. This can be achieved by inserting the event mark by means of a radio clock with an accurate time base. PATIENT ELECTRODES of step (e) and g) Adjust the rectangular pulse generator so that a 2. reverse the positive and negative record at least 30 s of the test waveform. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor. to P2. . Compliance is checked by the following test: The EQUIPMENT is arranged to record a signal from an ECG simulator or operate in the calibration mode for 24 h.1. for full disclosure).1 ± 0. the hard copy shall be ruled with 1.0 mm divisions along the horizontal (time) and the vertical (voltage) axis. 8 h ± 1 s and 23 h ± 1 s into the test.0 ms ± 0.5.0 ms ± 0.0 ± 0. *51.1 ms.1 ms. ELECTRODE k) For each channel. e) Reduce the pulse width to 0. measure the ruling in the three following distinct locations within an 7 s strip of the recording paper: in the upper left corner. then it is acceptable not to use grid rulings. a rise time ≤ 100 μ s and a repetition rate of 300 pulses/min is present across the 111 Ω resistor.5.2 mV pulses with a duration of 2.13 Hard copy grid standard Except as indicated below. insert an event mark on the recording. The rulings shall be accurate within ±2 % over the humidity conditions specified in 10. 8th. and 23rd hour of the report. f) For each channel. and record at least 30 s of this test waveform.05 mm. a mark at least 2 mm high is printed on the hard copy record at the same repetition frequency and same inter-pulse interval as the pulses input into the EQUIPMENT . Copying. in 16 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. with the positive PATIENT ELECTRODE connection for each channel connected to P1 and the negative ELECTRODE connection for each channel. reverse the positive and negative record at least 30 s of the test waveform.2) mV p-v sinusoidal signal is present across the 111 Ω resistor. networking. b) Adjust the sine wave generator so that a 10 Hz (2. h) Record at least 30 s connections in (g) and repeat the recording. as well as the reference ELECTRODE connection. Inspect the full disclosure report and verify that each event marks actual time of day within 30 s of the 1st.1 ± 0. *51. l) PATIENT ELECTRODES of step (j) and Confirm on playback that for every pulse. Every fifth ruling shall be accented.12 Timing accuracy The overall error during 24 h shall not exceed 30 s. At 1 h ± 1 s.

5 mm) 51. Connect all positive LEAD connections to point P1 and all negative LEAD connections to point P2. *51.5 mm).05 mV across Pl and P2. During the suspension of the scan. The VSS display device should be capable of displaying the ECG signals with minimum scaling factors of 25 mm/s horizontally and 10 mm/mV vertically.and fall-time of <1. a) b) c) d) e) f) There shall be available a minimum display of two channels of simultaneous ECG as a default. The signal source is adjusted to provide a train of rectangular pulses that have an amplitude of 1.). The system may allow the display of one or more channels to be turned off. a duration of 200 ms. networking.the middle.15 Temporal alignment When the amplifiers for all channels are set to the same frequency response limits. If the skew exceeds the above stated limit. Scaling factors exceeding these dimensions shall. Repeat this test for all playback speeds available in the scanning EQUIPMENT . 17 . the channel to channel skew shall. the spacing between every 10th ruling within a square of 20 mm × 20 mm shall be within ±2 % (±0. a rise. and distribution prohibited. d) Verify that a warning is printed or displayed by the PLAYBACK EQUIPMENT if the measured skew exceeds 20 ms (0.2 mm). c) Record at least 1 h of pulses on all the channels of the RECORDER . Interaction capabilities shall be provided to permit the operator to stop and to resume the scanning process at any point during the recorded period. then a suitable warning shall be included in the record to indicate that channel to channel temporal comparison is not advised. Print out the signal of each channel and print the signals of at least two channels at a time (or display them) with a resolution of 25 mm/s and 10 mm/mV. VSS displays shall meet the minimum frequency response and linearity requirements specified in this standard for the rest of the system.5. Verify that the skew of the rising and falling edges of the signal between each of the channels is less than 20 ms (0. and switch S5 is open.0 + 0. Compliance is checked by the following test: a) Connect the RECORDER to the test circuit of figure 104. b) If the RECORDER or the PLAYBACK EQUIPMENT has switchable amplifier filters. etc.14 Gain settings and switching The gain used shall be printed out on the printout. *51. except as indicated below. Copying. This applies to the whole system and all its individual component parts ( RECORDER . and in the lower right corner. Switches Sl and S2 are closed. retain the 5:2 aspect ratio within 10%. At each location. by default. such as separate windows or color differentiation. The maximum speed of the visual superimposition display shall be at least 60 times real time. adjust them such that all channels have the same frequency response. PLAYBACK EQUIPMENT .0 ms and a repetition rate of 1/s.5 mm (at 25 mm/s time axis scale). The system shall have a method for operator validation of beat classification including ectopy and artifact.5.5. as reflected in real time. hard copy printout capabilities shall be provided for examination of suspected arrhythmias or artifact. switch S3 is in position A. be less than ±20 ms or ±0. If abnormal beats are © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. Analog systems shall provide a calibration pulse on the printout. Make this measurement at three distinct points for each channel in the 1 h record. Compliance is checked by inspection of the printout.16 Special considerations for high speed superimposition display (optional) The visual superimposition system (VSS) shall meet the following requirements.

there shall not be a loss of all ECG signals if one of the other PATIENT ELECTRODE connections is lost. networking. the equipment shall have the capability for user verification of the acceptability of the electrode placement. 51. If the device supports a multi-channel lead system with independent ECG channels. If the device supports a dependant lead system such as a 3-wire 2-channel electrode system where one of the lead wires is utilized by both channels and that common lead wire is compromised. fixed lead definitions. the device shall be so constructed that the loss of any one PATIENT ELECTRODE connection will not result in the loss of all ECG signals.5. remove each electrode one at a time for at least 15 s. with particular consideration given to safety characteristics during use and during 18 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. Verification of electrode placement At the time of patient hookup. In this configuration. the exclusion criteria shall be disclosed in the physician’s guide. the connector shall be constructed so as to prevent insertion and electrical connection into a mains outlet or a detachable power cord. both channels will be compromised. . b) For devices that have multiple dependent leads. and any input signal can be visually displayed or printed or indication of signal quality for any signal can be visually displayed or printed.excluded from VSS display. In particular. as are other devices. This may be accomplished with any of a number of methodologies. the electrode lead wire connector shall be constructed in such a manner that the pins in the connector used to mate with the AECG recorder cannot contact ground or a possibly hazardous electrical potential. Compliance is checked by the following test: a) With the test setup of figure 104.17 Lead Definitions Ambulatory monitoring is not predicated upon specific. c) For devices that have multiple independent channels. Compliance can be checked by the following test: Compliance with the electrode placement verification mechanism requirement is determined by demonstrating that the acceptability of the lead placement can be confirmed. Safe electrode lead wire connectors In conformance with the referenced standard on disposable ECG electrodes (ANSI/AAMI EC12:2000/(R)2005). the following minimum lead requirements shall be met to allow the equipment to function over a wide range of clinical situations: Number of leads Devices that support configurations of one channel of ECG signal (usually used in long term monitoring) are exempt from this requirement. remove each non-common electrode one at a time for at least 15 s. d) Confirm that the sinusoidal signal is always observable in at least one channel during the entire recording. record a 2 mV p-v sinusoidal signal for at least 30 s. provided that the user can determine the acceptability of the lead placement and signal quality. Compliance can be checked by the following test: Compliance with the requirements of safe electrode lead wire connectors can be determined by inspection. Copying. and distribution prohibited. however.

any gaps in the recorded data shall be indicated in a consistent manner occupying the same space on the hard copy as if the real-time ECG had been recorded during that time. SECTION TEN – CONSTRUCTIONAL REQUIREMENTS The clauses and subclauses of this section of the General Standard apply except as follows: *56 Components and general assembly *56.g.20 Full disclosure (miniature displays) A claim of full disclosure capability requires that a hard copy record of all recorded data for 24 h be available.19 Patient event marks The major value of the AECG is the capability of correlating transient ECG events with patient symptoms. 51. Compliance is checked by inspection and verification.18 Baseline stability With the sensitivity set to maximum and all patient lead wires connected together with 25 KΩ resistors inserted in series with any or all of the leads. 51.5. 51. and time of day shall be present on each page of the record. Copying. Patient identification. date.attempted forced mating with connectors supplying hazardous energy (e. This permits more accurate correlation of brief symptoms with transient ECG phenomena than is possible with only time-of-day logging. A patient-activated marker shall be provided.5. Addition: © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 19 . Indications of patient-activated marks aligned to the coincident ECG are preferred SECTION NINE – ABNORMAL OPERATION AND FAULT CONDITIONS.7 Batteries c) Battery state Replacement: Digital AMBULATORY RECORDERS shall meet the following requirement: Means shall be provided to indicate when any INTERNAL ELECTRICAL POWER SOURCE is discharged to a degree where the EQUIPMENT is incapable of meeting the requirements of this Particular Standard. female end of a detachable power cord or power outlet).5 mm/s. Total time to print the 24 h disclosure shall be less than 2 h. and distribution prohibited. The hard copy record may be at compressed time and voltage scales.5. ENVIRONMENTAL TESTS The clauses and subclauses of this section of the General Standard apply. The voltage scale shall not be less than 1 mm/mV. and the time scale not less than 2. networking. the baseline drift shall not exceed 1 mV over 24 hours.

1 20 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.101 Monitoring time and retention of data *56.7. networking.201.05 m 3 6 Ch Rh IEC 947/01 Key 1 POWER SUPPLY CORD 2 Signal cable 3 Table made of insulating material 4 EQUIPMENT under test 5 6 7 PATIENT CABLE Load simulating the PATIENT (51 kΩ in parallel with 47 nF) Metal plate C h = 220 pF R h = 510 Ω (C h in series with R h simulates a hand. of monitoring for at least 24 h continuously.5 m 1m 0.101.101.3 m 0.101. .*56. c) Read the recorded data at the end of this period and verify that no change of the data is apparent.1 a).2 Data retention using volatile memory support shall be capable of retaining the stored information without any external power supply for at least 72 h after completion of the monitoring time.7. >1 m 0.3 m 2 EUT 1 4 1m 5 7 0.7. and distribution prohibited.1 Monitoring time a) CONTINUOUS RECORDERS SOURCE as specified by the shall be capable. POWER Compliance is checked by measurement.7.) Figure 101 – Test set-up for conductive emission test according to 36. with a fully charged INTERNAL ELECTRICAL manufacturer. b) Leave the EQUIPMENT isolated for 72 h after recording at a temperature of 25 °C and a humidity of 70 %. Copying. *56. RECORDERS Compliance is checked by: a) Perform the test as per the requirement of 56.

2 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.5 m 0.3 m 2 1m EUT 1 4 7 5 0. 21 . Copying. and distribution prohibited.202.1 and 36.>1 m 1m 0.05 m 6 IEC 948/01 Key 1 POWER SUPPLY CORD 2 Signal cable 3 Table made of insulating material 4 EQUIPMENT under test 5 6 7 PATIENT CABLE Load simulating the PATIENT (51 kΩ in parallel with 47 nF) Metal plate Figure 102 – Test set-up for radiated emission and radiated immunity test according to 36.3 m 3 0.201. networking.

96 ms

6 mV

IEC 949/01

Figure 103 – Test signal for input dynamic range test according to 51.5.1

S4

300 mV d.c.
adjustable

S1
4,7 nF
100 kΩ
0,1 % P5 S2

b
P4

P1
0,62 MΩ

a

S3

S5

100 Ω
0,1 %

RECORDER

P2

P3
51 kΩ
47 nF

P6
RL
IEC 950/01

Figure 104 – General test circuit for 51.5

22

© 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007

Single user license only. Copying, networking, and distribution prohibited.

Driven shield around entire test configuration
Sb
± 300 mV
Driven shield
connected at
this point

S1
Sa

R
C1
100 pF

C

Input
S2

R
Ct

C

Line
freq.

Input
S3

RECORDER

R

C

Input
Sn

R

Input
Repeat for each
input connection

C

Reference

R
Cx
C

Driven shield

Foil wrapped
around recorder

(Stray)
Earth reference

R = 51 kΩ
C = 47 nF

IEC 951/01

Figure 105 – Test circuit for common mode rejection according to 51.5.3

© 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007

Single user license only. Copying, networking, and distribution prohibited.

23

100 kΩ

P1

1 kΩ
2,0 V p-v
10 Hz

111 Ω

Recorder

0,02-2,0 V
0.02-2.5 V
100
300 ppm
ppm
P2
Reference

IEC

952/01

Figure 106 – Test circuit for pacemaker pulse tolerance according to 51.5.11
(ppm = pulse per minute)

24

© 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007

Single user license only. Copying, networking, and distribution prohibited.

and distribution prohibited. including essential performance. networking. pp. ANSI/AAMI EC57:1998/(R)2003. Recommendations for standards of instrumentation and practice in the use of ambulatory electrocardiography. L. of electrocardiac monitoring equipment IEC 60529:1989. Copying. 25 . Medical electrical equipment – Part 1: General requirements for safety Amendment 1 (1991) Amendment 2 (1995) IEC 60601-2-25:1993. Degree of protection provided by enclosures (IP code) SHEFFIELD. Medical electrical equipment – Part 2-25: Particular requirements for the safety of electrocardiographs Amendment 1 (1999) IEC 60601-2-27:2005. (AHA special report from the task force of the Committee on Electrocardiography and Cardiac Electrophysiology of the Council on Clinical Cardiology).. 1985. Medical electrical equipment – Part 2-27: Particular requirements for the safety. vol. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 71. 626A-636A. Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms IEC 60601-1:1988. et al.T.Appendix L (normative) References – Publications mentioned in this standard ASSOCIATION FOR THE ADVANCEMENT OF MEDICAL INSTRUMENTATION. Circulation.

2 Radiated radio-frequency electromagnetic fields Some PATIENT s may work in an environment that has unusually high electromagnetic fields. RECORDERS 21. There are no different requirements specified for tape type and solid-state memory type of since both types of equipment are used in the same environment.101. these PATIENT s should be advised by the physician not to expose these EQUIPMENT s to such high fields. Copying. 36 Electromagnetic compatibility EMC of medical EQUIPMENT s is a growing concern. Furthermore.202. The collateral standard IEC 60601-1-2 reflects this concern and provides general guidance on EMC requirements.Annex AA (informative) Guidance and rationale This annex provides a concise rationale for the important requirements of this Standard. It is intended for those who are familiar with the subject of the Standard but who have not participated in its development.5 Other standards relating to small parts connected to the PATIENT have in the past specified a timber density of ≥700 kg/m 3 .202.101 50. The fact that AMBULATORY RECORDER s are coupled to the PATIENT and have to operate accurately in a variety of locations where the electromagnetic environment is not known or is widely variable. The 10. However. A lower density is therefore specified in this standard. 10 Environmental conditions An extended range for temperature and humidity are required since RECORDERS may be used outside medically used rooms. it is believed that a rationale for the present requirements will facilitate any future revision of the Standard necessitated by these developments.4 Hz modulation frequency was selected because it is within the EQUIPMENT s’ pass band and is near to the frequency of the maximum energy of the QRS complex. 50 Accuracy of operating data 50. 36.2 Performance reporting requirements © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. An understanding of the reasons for the main requirements is considered to be essential for the proper application of the Standard. as clinical practice and technology change. and distribution prohibited. networking. the working group is aware that this equipment is in constant contact with PATIENT s and may be subjected to larger electrostatic discharges. . To avoid improper recordings.101. Policies relating to the conservation of hardwoods have made this density of timber unobtainable. make it necessary to expand the general requirements of IEC 60601-1-2 in order to provide reasonable assurance that the equipment will always perform well and safely.1 26 Automated analysis Use of standard databases and 50. 36. The requirements specified are intended to cover the majority of environmental conditions likely to be encountered in practical use. Test house experience has shown that such timber is entirely satisfactory for the test.1 Electrostatic discharge The value of 6 kV specified in this Standard was selected because it is a reasonable value that is in common usage.

S. In the early 1990s. ST segment measurement algorithms and heart rate and heart rate variability calculations. EC57 was reaffirmed without changes leading to its current designation as ANSI/AAMI EC57:1998/(R)2003. In the early 1990s. The AAMI ECG committee decided (partially because membership of WG 01 and WG 02 overlaps) that WG 01 would lead the development of reporting requirements for ST segment measurement algorithms and heart rate and heart rate variability calculations. replaced ECAR with ANSI/AAMI EC57: Testing and reporting performance results of cardiac rhythm and ST segment measurement algorithms. EC/WG 01 was chartered to develop a performance standard for ambulatory electrocardiographs that perform ventricular arrhythmia detection and.3 of IEC 60601-2-47. These were 1) whether to adopt the IEC 60601-2-47 standard (with or without US deviations). optionally. that document was approved as an AAMI recommended practice (AAMI ECAR:1987) with the same title. WG 01 adopted ECAR as its reporting prototype. A number of other organizations have also done so. Many of the reporting requirements found in section 50 of IEC 60601-2-47:2001 appear to derive from AAMI EC38:1998. ST segment measurement algorithms.101. 27 . heart rate and heart rate variability calculations.The U. In April 1987. and 2) how to deal with the fact that EC57’s reporting requirements are more complete than those found in IEC 60601-2-47. deviations in this area list ANSI/AAMI EC57:1998/(R)2003 as an additional normative reference and delete IEC 60601-2-47’s addition of sections 50. EC/WG 02 began extending ECAR to cover other measurement of interest. In 2003. but recognized that ECAR did not cover WG 01’s other signal of interest. networking. WG 01 faced two primary issues with respect to revising EC38:1998. an IEC working group began writing a standard covering safety and essential performance of ambulatory electrocardiographic systems. AAMI/EC WG 02 (Arrhythmia Monitoring Working Group) of AAMI ECG Committee published a recommended practice titled “Recommended Practice for Testing and Reporting Performance Results of Ventricular Arrhythmia Detection Algorithms”. This led WG 01 to include requirements for reporting of results for both ventricular arrhythmia and ST segment measurement algorithms. WG 01 decided to adopt IEC 60601-2-47:2001 with US deviations (primarily relating to reporting of performance results). WG 02 would assist WG 01 with this with the stated intention of publishing a revised version of ECAR after AAMI EC38 was completed. While EC38 was being finalized. Copying. the ANSI/AAMI standards include all reporting requirements from the start of section 50. In the late 1990s. deviations from IEC 60601-2-47 related to reporting of algorithm performance results are based upon the timing of the publication of the standard for ambulatory electrocardiographs (ANSI/AAMI EC38:1998) and the recommended practice for reporting of algorithm performance results (AAMI ECAR and ANSI/AAMI EC57:1998/(R)2003) with respect to each other. and distribution prohibited.101 through the end or 50.S. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. The standardized reporting mechanism described in ECAR:1987 quickly gained widespread recognition as a significant advancement in reporting of performance results. The reporting requirements incorporated in IEC 60601-2-47 were handled as an “addition” to the second edition of the general standard (IEC 60601).101.2. which was published as ANSI/AAMI EC38:1998. At least two groups within AAMI. ST segment analysis. the Apnea Monitoring Committee and EC/WG 01 (Ambulatory Electrocardiographs Working Group) immediately adopted ECAR as their prototype for reporting of performance results related to physiological data. The way these requirements were incorporated into IEC 60601-2-47 will become significant later. WG 02 also did some additional work on the resulting document and. in 1998. WG 02 incorporated the EC38 reporting requirements into ECAR. The U.101. The result is that ANSI/AAMI EC57:1998’s reporting requirements are more complete than those found in ANSI/AAMI EC38:1998. To be specific. but not those found in section 50.

networking. to report performance results. A 6 mV p-v.ELECTRODE impedance levels over the frequency range of the ECG signal. bikers) may often have resting rates below 50 bpm. measuring systems should have sufficiently high input impedance that practically all subjects will be measured without significant errors. 51. Industry is able to meet these requirements easily. 50. bradycardia. Since long-distance athletes (runners. hence a ±3 mV input dynamic range and slew capability of 125 mV/s are quite sufficient. so that cardiac pathology is not falsely diagnosed in such cases. an EQUIPMENT whose 10 ended input impedance magnitude is 9.).5. the cited studies are still relevant for worst-case limits.ELECTRODE impedances ELECTRODE application.7 Hz sine wave has a maximum rate of change of 125.5. Hence. which is the most complete set of reporting requirements. It is essential that the RECORDER s perform adequately in the presence of substantial d. the Hz single- Common mode rejection The particular method of specifying and measuring common mode rejection (CMR) chosen for this standard produces a worse than normal configuration of capacitance to ground in relation to capacitance to PATIENT from the RECORDER .c. The CMR requirement of 60 dB at line frequency is fairly conservative. The test decrease with increasing frequency and with method simulates the frequency dependent drop in by means of a 4. At impedance of this combination is about 610 kΩ. The specification of ±300 mV offset tolerance is sufficient for the ELECTRODE polarization encountered. This requirement originally arose from the need to deal with large ELECTRODE polarization voltages. The use of modern reduced impedance ELECTRODE s will decrease further the small number of subjects whose excessive ELECTRODE -to-skin impedance causes measurement error. 6.101. The line frequency harmonic measurements are included because of power line waveform distortion that 28 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.7 nF capacitor connected in parallel with a 0. Copying. and distribution prohibited. it is important to have the capability for OPERATOR selected parameters.50. offset voltages.1 Incorrect output Input dynamic range interpretations do not ordinarily involve analysis of the QRS in fine morphological details.1 and 50.66 mV/s.3 time after impedance 10 Hz.5. Skin-to. ECG The maximum rate of change of a sine wave equals the p-v amplitude times the frequency times π. Although continued development of pregelled ELECTRODE s has resulted in even lower average impedance levels. If conventional ECG ELECTRODE s are to be used.101.3 Minimum physician report requirement It is important to flag episodes of tachycardia.101.2 Input impedance The input impedance is primarily set by effective skin-to. 51 Protection against hazardous output 51.62 MΩ resistor.2 to the general standard.55 MΩ or greater will pass this test. ectopy and ST segment shifts and to bring these episodes to the attention of the physician taking care of the PATIENT . The capability for OPERATOR selected parameters may also be useful for ST segment analysis. swimmers. The result is that the reader must use EC57. 51. While less than the 320 mV/s slew rate specified for diagnostic ECG s and cardiac monitors. In actual use where the dominant capacitance is that between the RECORDER monitor and the PATIENT . significantly higher CMR performance can be expected. as use of older style ELECTRODE s persists.5 51. this ambulatory ECG requirement does exceed the 75 mV/s recommended by the 1985 AHA Report (Sheffield et al. .

It is desirable to achieve good common mode rejection at frequencies other than the SUPPLY MAINS frequency. The RECORDER is encapsulated in earth-grounded foil in order to define and stabilize the capacitance of the RECORDER to earth. 51.1 mV) shift can indicate the difference between normal and ischemic repolarization. where a 1 mm (0. the greater the difficulty in meeting the requirements set forth with this particular test methodology. and distribution prohibited. as well as high-input impedance and common mode rejection. The ability is critical to the accurate measurement of ST segment levels. 51.5 Gain stability In order to guarantee consistent interpretation. digital ambulatory ECG systems. Thus analog systems employing removable media conventionally use calibration pulses to adjust playback amplifier gain. alleviate some of the noise problems.5. 1985). On the other hand. Changes in the ECG signal that do not stem from physiological or pathophysiological changes have to be minimized. Additional settings are not needed to guarantee safety and efficacy. 29 . However. this test mostly checks the (differential mode) rejection of such a notch filter. In this standard. diagnosable signal. Most manufacturers provide guidelines for correct techniques in measuring ECG . however. which helps cancel the common mode noise from the signal sensed at the ELECTRODE s. The higher the resulting capacitance of the RECORDER to the foil wrap. The only remaining variable that will influence the test is the physical design of the particular RECORDER . thereby assuring needed overall system gain accuracy. can generally be traced to external interference (EMI). ST segment analysis has been a major application of visual superposition systems and is becoming increasingly implemented in more automated. the guarding on the input test components and PATIENT CABLE reduces the difficulty in meeting this performance specification. the thickness of the insulated case).6 Amplitude calibration The standardizing voltage ensures the ability of an analog system to reproduce a signal relative to an absolute reference. The limits specified here represent a consensus on achievable levels established in practice. Analog media are inherently more susceptible to gain variances than digital media. The input test components and the PATIENT CABLE are guarded by the driven shield that eliminates the effect of stray capacitance to earth ground from those components. which fixes the size of the RECORDER and also sets the spacing between the internal circuitry and the external foil wrap (that is.4 Gain accuracy The reference gain setting of 10 mm/mV reflects a well-established convention (AHA recommendations. 51.5. gain stability is especially important in ambulatory monitoring equipment where the typical monitoring period is 24 h or more.5.5. Shefffield.can occur because of discontinuous loads from items such as SCR controllers and electronic equipment power supplies with capacitor input filters. the entire test set-up is enclosed in an earth referenced shield in order to stabilize the Cx stray capacitance. The driven right leg. Shielded cables. further reduces the noise.7 System noise Noise in electrocardiographic records is one of the most persistent detriments to a clean. PATIENT movement (myographic signals) or poor technique in ELECTRODE application or routing of cables. In order to check the common mode rejection of the EQUIPMENT s circuitry. This problem. the system output at all available gain settings has to be guaranteed to be reasonably close to that of an ideal system. it is necessary to disable any SUPPLY MAINS frequency notch filter. 51. The limits of 5 % error represents a reasonable standard for performance. Furthermore. Copying. networking. Otherwise. the © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.

often diagnostically important.5.05 Hz high pass filter has been considered acceptable for high pass filtering of an ECG signal such that the ST segment was not distorted.5. However. If the frequency response requirement of IEC 60601-2-47 is retained.2:pp730-739). Bailey et al. The slope requirement of 0. 0. 30 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. The baseline offset allowed following the impulse represents the droop that can be expected from a 0. and distribution prohibited. This low frequency specification has been accepted both for ambulatory and diagnostic ECG signal acquisition.67 Hz can be used to filter ECG signals.10 Minimum feature size Clear responses to 50 μV signals increase the likelihood that low-amplitude P waves.67 Hz low frequency specifications in the 1990 AHA recommendations. The use of properly designed digital filter with high pass cutoff frequency at 0.30 mV/s translates at standard gain and speed to a change of 0. The level specified is quite sufficient for diagnostic purposes and economically feasible in practice. Circulation 1990:vol. used these data to justify the 0. We feel that this approach is not dropping a requirement. This specification is based on that in the draft IEC ECG standard.67 Hz. then ECG baselines will be corrupted with much higher levels of low frequency artifacts such as respiration induced baseline wander and drift. For this reason. but is merely separating the amplitude and frequency response effects of filters on the ST segment. The low frequency response of 0. the adequate frequency response of filters was determined based on the characteristics of analog hardware filters. Since 1991. a high pass filter cutoff frequency of 0. 51. 51.05 Hz highpass filter. but are not guaranteed to preserve the ST segment. Furthermore. Thus.11 Function in the presence of pacemaker pulse PATIENT s having large pacemaker pulse amplitudes at the body surface are prevalent in the population receiving ambulatory ECG s for diagnostic purposes. can be discerned. the AAMI EC11 standard for diagnostic ECG has not constrained amplitude response below 0. These studies indicate that 44 BPM encompass more than 99% of adult heart rates with intra-individual RR interval variation of less than 126 ms.67 Hz and has controlled phase response with the 3 mV rectangle waveform test. Bailey et al. digital signal processing methods have enabled design of digital filters that allow more aggressive filtering of baseline wander without distorting the ST segment. In the case of linear or zero phase filters.maximum allowable noise was harmonized with the corresponding requirement in the draft IEC ECG standard.3 mm across a typical ST interval of 100 ms and would not be clinically significant. In this case. a lower bound of 40 BPM (0. 51. "Recommendations for Standardization and Specifications in Automated Electrocardiography: Bandwidth and Digital Signal Processing".81:no. 51.1 mV/s (cf. devices with adequate phase and frequency response should be able to pass both the sinusoidal test (test b) and triangle test (test c). 90% of the time.67 Hz is based upon heart rate data from the Framingham Heart Study and Simonson’s studies. The more relaxed levels of droop are appropriate for single pole filters that have a cutoff frequency at 0. Copying. British Heart Journal. the phase response of a 1 pole 0. "Artefactual ST Segment Abnormalities Due to Electrocardiograph Design".9 Frequency response Historically.8 Multichannel crosstalk The maximum level of crosstalk is determined by the requirements of accurate diagnosis and the incremental costs of noise suppression. networking. 1985:vol.5. particularly as a typical QRS complex has an impulse value closer to 0.67 Hz have been accepted since 1985 (Taylor and Vincent.67 Hz) exists for 99% of adults.30 mV/s).54:pp121-128. the ECG recorded by EQUIPMENT s that are claimed to be suitable for use with such PATIENT s should not be unduly distorted by the presence of the pacemaker pulse signal. .5.

51. It can also be used to identify displacements of ST segments (Holter. drug administration.15 OPERATOR with a means of Temporal alignment Clinical evaluation of the ECG recorded requires viewing the ECG complex in different planes. Concurrent analysis of two (or more) ECG channels requires that the channel-to-channel skew be sufficiently small so as not to influence the clinical interpretation of the ECG . Copying. A NOMINAL scaling factor of 10 mm/mV has been traditionally used. 31 . networking.5 mm corresponds to a maximum temporal skew of ±20 ms (at a scale of 25 mm/s). Inclusion of the ECG calibration pulse in the display provides the verification of the reproduced ECG amplitude. 40 mm/mV and 2. 51. such as pacemaker pulses. This tolerance accounts for practical measurement errors and also for variations in the recording and PLAYBACK EQUIPMENT due to alignment of magnetic recording heads due to tape tracking and/or analog-to-digital conversion skew. It can be used to review a contiguous sequence of QRS complexes (from a few minutes to many h) to search manually for arrhythmias (especially atrial arrhythmias) either not detected or not correctly identified by the automated analysis.5. Since the recorded ECG is reviewed retrospectively. The specification of a maximum ±2 % error provides sufficient accuracy for clinical measurements and is attainable.14 Gain settings and switching Variations in the recorded amplitude of a PATIENT s ECG (due to physiological factors or LEAD placement) warrant the selection of different scale factors. The skew error of ±0.5. 1957). The additional gain settings of 5 mm/mV and 20 mm/mV conform to both IEC and AHA recommendations. A common example of such usage is VSS of members of a QRS form family. Other gain settings (for example. the recording EQUIPMENT and system have to provide a mechanism for accurately indicating the actual occurrence time along with the ECG signals on both the display and printout. Accentuation of every fifth ruling simplifies time and voltage measurements. 51. Changes in humidity will result in expansion and contraction of the paper fibers. and distribution prohibited. it can be used to verify that all members of a select data subset. laser printers). The cumulative accuracy requirement of ±30 s over 24 h ensures that the difference between the actual occurrence time and the recorded occurrence time is sufficiently small to allow clinical correlation and interpretation of the concomitant ECG complexes. physical activity) and the PATIENT s ECG complexes is essential to clinical interpretation. 51. 51. are sometimes only discernible in one plane.If the manufacturer claims that the EQUIPMENT is capable of recording pacemaker pulses. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.12 Timing accuracy The correlation of the occurrence time of an external event (for example. tests with four different pulses have to be done to cover the range of possible pacemaker pulses on the surface ECG .5.5. Continuous gain control is generally not desirable. Certain features.5.5 mm/mV) may be provided at the option of the manufacturer. Both measurement error and humidity are accounted for in the tolerance specifications of ±2 %. Additionally. The grid pattern can be pre-printed on the output paper or may be printed by the same printing mechanism used to print the ECG (for example. are authentic members of this subset. obtained during automated analysis of the data.13 Hard copy grid standard A rectilinear co-ordinate grid with a time axis of 25 mm/s and a voltage axis of 10 mm/mV has been traditional in electrocardiography.16 Special considerations for high speed superimposition display (optional) Visual Superimposition Scanning (VSS) can be utilized in at least two different ways. presence of symptoms.

some fatal. 51. Even if performed correctly. permits detection of QRS axis shifts due to transient intraventricular conduction defects. Ideally.18 Baseline stability The 10 μV/s requirement essentially allows less than 100 μV drift in a typical 10 s strip output. the electrode placement might not produce the desired waveforms or results due to the large number of variables involved. d) The need to observe the usually fine details of atrial activity requires signal quality to be as good as elsewhere in the system. and distribution prohibited. as both are elicited during patient activity. More ECG channels (beyond two) can in certain clinical situations yield additional clinical information. b) Due to the variation in sizes of CRT displays that are commonly available and the ease of interchanging one CRT for another. the lead wire connector pins must not be permitted to contact a possibly hazardous potential. or a conductive surface that may be at ground potential. speeds greater than 240 times real time are very useful in verification of the members of a specific subset. Safe electrode lead wire connectors Several incidents. displacements of ST segments may be apparent in only one of the two leads. for normal usage it is believed to be too slow and cumbersome and is not recommended. c) While systems may allow scanning speeds below 60 times real time. 51. arising from electrode lead wires attached to the patient having exposed male pins at the other end that may come into contact with a power source or may inadvertently be inserted into an inappropriate outlet or a detachable power cord. Copying. which might limit innovation. A major benefit of VSS display is the ability to survey atrial and atrial ventricular activity. Speeds greater than 240 times real time are not recommended for manual arrhythmia searches due to the short time the data are displayed. e) Periods of artifact and abnormal events are frequently correlated. As the vectors of atrial depolarization are directed differently from those of ventricular depolarization. and permits distinction between ventricular ectopic and supraventricular aberrant complexes by viewing the QRS complex in more than one plane. however. A 32 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.5. Such a drift would not produce any misinterpretation of the ECG due to measurement errors. This requirement provides redundancy and a record of at least one channel in the event of a single channel malfunction. Verification of electrode placement An erroneous or defective lead attachment can render an ambulatory ECG monitoring procedure useless.5. . though the standard does not impose a specific design. devices should support two or more independent channels of ECG signal. networking. two channels become necessary to capture both atrial and ventricular activities. The requirement for the capability to verify the hookup greatly increases the chances for a successful procedure.17 Lead definitions Number of leads The requirement for a minimum of two channels is taken from AHA recommendations. two different channels are needed to discriminate between normal and abnormal beats. This can be achieved by recessing the pins inside a surrounding insulating shield. thereby compromising patient insulation.a) Frequently. Furthermore. have been reported in recent years. To ensure patient safety. Dual channel recording also permits better assessment of QRST changes that result from body position shifts. It also permits improved detection of P-wave and ST-segment shifts that may be best detected along different lead axes. it is not practical to require that a standard size signal be displayed on all displays.

This time-span permits detection of most episodes of intermittent arrhythmias during waking and sleeping phases with recognition of temporal variability in frequency while providing adequate protection from the effects of intermittent recorder malfunction. For most clinical uses. Indications of the actual time of day.5. Copying. 51. As patient event indications may provide clinical information. 56.101 56. The total drift rate of 1 mV ensures that the baseline will remain fairly close to the center of the output display over long periods of time without constant operator adjustment. Events that occur frequently may be detected with short recording periods whereas those that occur infrequently or rarely may require prolonged recording. of the patient identity. 56. and distribution prohibited.101. it is desirable that these also be indicated on the full-disclosure report. This period takes into account the circadian variability of the processes underlying the cardiac activity. The minimum scale factors of 2. it is then necessary that the status of any INTERNAL ELECTRICAL POWER SOURCE be indicated to the OPERATOR who will then be able to substitute or charge the battery prior to installing the equipment on a PATIENT . Because of the particular application of the RECORDER (monitoring 24 h or more of the heart activity of a patient without the presence of specialized personnel) it is necessary to check the status of any internal electrical power source during the monitoring period in order to avoid processing and analyzing (either in real-time or off-line) corrupted or inaccurate data. and of any lapses in the ECG signal acquisition are necessary to form a complete account of the procedure.2 Data retention The requirements for 72 h data retention are based on the assumption that a stored recording should be kept for at least the duration of a week-end which is a common case which occurs in practice.60 s period is considered to represent a typical duration of a single ECG line across a full disclosure printout where drift of subsequent lines over each other can make interpretation difficult. If the function of any INTERNAL ELECTRICAL POWER SOURCE is necessary to ensure safe operation according to this Particular Standard (which includes accuracy of data and of the automated analysis).5 mm/s and 1 mm/mV reduce the printout record size while providing a clinically useful display of the ECG. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 51.20 Full disclosure (miniature displays) A full-disclosure display of the entire ECG record provides the physician with a means to scan the entire record as an aid in the physician’s evaluation. 33 . 56.1 Monitoring time and data retention Monitoring time The minimal time for ambulatory ECG recording varies with the indication for the test. a minimum recording period of 24 consecutive hours is recommended. networking. volatile memory back-up battery).19 Patient event marks No additional rationale is provided for this subclause.7 Batteries Some equipment makes use of more than one INTERNAL ELECTRICAL POWER SOURCE depending on the function to be supplied (for example.101.5.

HOLTER.) Circulation. et al.) Circulation. pp. Annals of Noninvasive Electrocardiology. 691–694. Ann NY Acad Sci. (AHA special report by a task force of the Council on Clinical Cardiology. JJ and SCHMITT. and clinical use. networking. MALIK. et al. Ann NY Acad Sci. KNOEBEL. The development of test methods for disposable ECG electrodes. JAGER. (AHA special report by an ad hoc writing group of the Committee on Electrocardiography and Cardiac Electrophysiology of the Council on Clinical Cardiology. Framingham Heart Study. Instrumentation and practice standards for electrocardiographic monitoring in special care units. et al. vol. EFRON. et al. 1996. JAGER. Journal of Ambulatory Monitoring 1(2): 171-176. p. Radioelectrocardiography: A new technique for cardiovascular studies. 79. MERRI. IEEE Computer Society Press: Los Alamitos. 1990. CA. N. April 1979. JAGER. HV. P. 11. NJ. 1989. M. 76. (Task force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology. 1957. 519. IEEE Trans Biomed Eng. UBTL TR 1605-005. HOLTER. RJ and LEVY. AS and PIPBERGER. vol. 1989. 1988. 170. F. AA. vol. 1979. BERSON. vol. Skin-electrode impedance problems in electrocardiography. Use of the ‘bootstrap’ to assess the robustness of the performance statistics of an arrhythmia detector. 1992. vol. 223-74-5253. JJ. New method for heart studies. BD. physiological interpretation. Computers in Cardiology. Part 1. GARRISON. pp. pp. 1992. 509. vol. p. IEEE Computer Society Press: Los Alamitos. Electrode impedance and voltage offset as they affect efficiency and accuracy of VCG and ECG measurement. et al. and distribution prohibited. 464–471. 1991. F. Bootstrap methods: another look at the jackknife. Recommendations for standardization and specifications in automated electrocardiography: Bandwidth and digital signal processing. pp. 17. CA. 1970. OH. vol. NJ. Computers in Cardiology. 1214–1220. 1990. standards of measurement. FDA Contract No.) Circulation. p. Circulation. Heart rate variability. 34 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. 93:1043–1065. 1961. B. Ambulatory ECG monitoring for the detection of ischemic events and ventricular arrhythmia: Differential utility of event recording versus conventional monitoring systems. P. 206–215. (American Heart Association Task Force). 103–112. 65. Analysis of transient ST changes during ambulatory monitoring using the Karhunen–Loeve transform. 453–456. OH. Salt Lake City: UBTL. JJ and SCHMITT. Systematic and random variations of ECG electrode system impedance. . NEARING. F. 7. et al. Final Report. D. pp.. 1996. SCHOENBERG. et al. 67A. 81. Guidelines for ambulatory electrocardiography. Annals of Statistics. Personal communication. et al. 369–372. 1–26. ALMASI. pp. vol. 1988. JACC. et al. Amer Heart J. et al. MIRVIS. 79. BAILEY. Copying. Science. Special report of joint ACC/AHA task force. Performance measures for algorithms to detect transient ischemic ST segment changes. 1968. vol.References – Informative ALBRECHT. M. New York: North-Holland Publishing Co. SB. Sampling frequency of the electrocardiogram for spectral analysis of the heart rate variability. DM. 1. 99–106. (American College of Cardiology Subcommittee). Analysis of transient ST changes during ambulatory monitoring. In Proceedings of the XIth International Vectocardiography Symposium. pp. and FISCH. pp. pp. DEEDWANIA. ALMASI. 730. CA. C. 913–923. pp. p. vol. Frequency response characteristics required for detection of T-wave alternans during ambulatory ECG monitoring. 134. Computers in Cardiology. IEEE Computer Society Press: Los Alamitos. 2. vol. et al. 1970. et al. et al.

p. 1949. 1987. Differentiation between normal and abnormal in electrocardiography. 34. 694. V. MS. St. © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only. p. Comparison of amplitude modulated (direct) and frequency-modulated ambulatory techniques for recording ischemic electrocardiographic changes. SIMONSON. 1966. et al. Circulation. Mosby Co. Louis: C. Copying. 35 . E. SIMONSON. p. Skin electrode impedance and its effect on recording cardiac potentials. E. and distribution prohibited. 158. Variability of the electrocardiogram in normal young men. vol. 60:895–900.SHOOK. et al. 407. Am Heart. Am J Cardiol. SPACH. networking. vol. 1961.. 38. et al.

1.12.1.7 60601-1 I NTERNAL ELECTRICAL POWER SOURCE I NTERNALLY POWERED EQUIPMENT 2.12.1. and distribution prohibited.5 60601-2-47 60601-2-47 60601-1 C ONTINUOUS 2. .4.INDEX OF DEFINED TERMS Defined Terms Clause Standard A CCESSIBLE PART A CCOMPANYING DOCUMENTS A MBULATORY ELECTROCARDIOGRAPHIC SYSTEM ( EQUIPMENT ) A MBULATORY RECORDER A PPLIED PART 2.12.104 2.10 60601-1 60601-1 60601-1 60601-1 T YPE T YPE T YPE 2.1.9 2.17 60601-1 60601-2-47 60601-2-47 60601-2-47 60601-1 60601-1 S AFETY HAZARD S IGNAL INPUT PART S IGNAL OUTPUT PART S UPPLY MAINS 2.24 2.4 60601-1 60601-1 2.1.109 2.1.101 2.110 60601-2-47 E LECTROCARDIOGRAM ( ECG ) E LECTRODE E NCLOSURE 2.26 60601-1 60601-1 60601-1 RECORDER APPLIED PART CONDUCTOR B APPLIED PART BF APPLIED PART CF APPLIED PART ___________ 36 © 2008 Association for the Advancement of Medical Instrumentation „ AAMI EC38:2007 Single user license only.1.29 60601-1 60601-1 L EAD L EAD WIRE L EAKAGE CURRENT 2.6 2.6.2 2.107 60601-1 60601-2-47 O PERATOR 2.25 2.2.7.6 60601-2-25 60601-2-25 60601-1 F.3 60601-2-25 60601-2-47 60601-1 M AINS VOLTAGE N EUTRAL ELECTRODE 2. networking. Copying.12.1.18 2.1.4 2.19 2.106 2.108 2.102 2.103 2.103 2.17 60601-1 P ATIENT P ATIENT CABLE P ATIENT ELECTRODE P LAYBACK EQUIPMENT P OTENTIAL EQUALIZATION P OWER SUPPLY CORD 2.22 2.5.18 2.1.1.101 2.TYPE 2.