See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/7817049

Digital recording of surgical procedures using a personal computer

Article  in  European Journal of Obstetrics & Gynecology and Reproductive Biology · July 2005
DOI: 10.1016/j.ejogrb.2004.11.003 · Source: PubMed

CITATIONS READS

10 763

6 authors, including:

Adam Magos Ioannis Kosmas

Royal Free London NHS Foundation Trust University of Ioannina
309 PUBLICATIONS   6,738 CITATIONS    74 PUBLICATIONS   778 CITATIONS   

SEE PROFILE SEE PROFILE

Malini Sharma Alex Taylor

Bhavan’s College The Royal Bournemouth and Christchurch Hospitals NHS Foundation Trust
12 PUBLICATIONS   197 CITATIONS    16 PUBLICATIONS   390 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Laparoscopy View project

Uterine Rupture in Pregnancy: The URIDA (Uterine Rupture International Data Acquisition) Study View project

All content following this page was uploaded by Adam Magos on 14 January 2016.

The user has requested enhancement of the downloaded file.

 European Journal of Obstetrics & Gynecology and
Reproductive Biology 120 (2005) 206–209
www.elsevier.com/locate/ejogrb

Digital recording of surgical procedures using a personal computer

Adam Magos*, Ioannis Kosmas, Malini Sharma, Lucie Buck,
Lynne Chapman, Alex Taylor
Minimally Invasive Therapy Unit and Endoscopy Training Centre, University Department of Obstetrics and Gynaecology,
Royal Free Hospital, Pond Street, Hampstead, London NW3 2QG, UK
Received 25 July 2004; accepted 11 November 2004

Abstract

Objective: To develop a system for recording surgical procedure digitally using a personal computer with real-time compression of the video
signal.
Study design: We built the system around a modern personal computer with a large hard disk to allow recording of over 250 h of continuous
surgery. Digital capture from the camera was achieved using a standard external analogue-digital converter linked to the computer via a
firewire cable. The software for capturing, compressing and editing movie files were obtained free of charge from the internet. The optimal
settings for the software was determined.
Results: We have successfully used this system to record over 100 major and minor hysteroscopic, laparoscopic, vaginal and open
gynaecological. Despite compression, the quality of the movies was judged to be very good and still images excellent. The recordings could be
integrated in to standard presentation. Still pictures could be printed to provide hard copies for patients and medical notes, and movies burnt on
to CDs or DVDs.
Conclusions: A digital recording system built around a standard personal computer is relatively cheap, versatile and has a huge capacity to
record surgical procedures.
# 2004 Elsevier Ireland Ltd. All rights reserved.

Keywords: Digital recording; Computer; Surgery

1. Introduction lung resection, coronary artery bypass surgery), and even

Minimal access surgery is becoming routine across many
specialties, and has been heralded as ‘‘the most important
revolution in surgical technique since the early 1900s’’ [1].
Numerically the most dramatic example of this change in
surgical practice has been laparoscopic cholecystectomy
which, since its first description in 1987, has replaced open
surgery as the standard of care [2,3]. The list of endoscopic
procedures is forever increasing and embraces not only
gynaecology (e.g. hysterectomy, myomectomy, adnexect-
omy, pelvic floor reconstruction), but many other surgical
specialities such as general surgery (e.g. appendicetomy,
colon resection, Nissan fundoplication), urology (e.g.
nephrectomy, adrenalectomy), cardiothoracic surgery (e.g.
* Corresponding author. Tel.: +44 20 7431 1321; fax: +44 20 7431 1321.
E-mail address: a.magos@rfc.ucl.ac.uk (A. Magos).
plastic surgery (e.g. breast reconstruction) [3]. Develop-
ments in robotics and computer enhancement have been
predicted to extend the role for this type of surgery [4].
A key component in the development of endoscopic
surgery has been advances in miniature, high-resolution
cameras which allow the surgical team to visualise the
procedure on a colour monitor. This also makes it possible
for surgeons to record their surgery as documentation of the
procedure which can be used for teaching, research, audit
and patient education. Currently, VHS video recording is the
commonest media used, but digital recording is now a
possibility offering superior quality and convenience.
Several instrument manufacturers sell camera systems
which include the facility for digital recording. These
systems are expensive and have limited capacity for
continuous recording. We describe a video capture system
which is based on a standard personal computer (PC) and is

0301-2115/$ – see front matter # 2004 Elsevier Ireland Ltd. All rights reserved.
doi:10.1016/j.ejogrb.2004.11.003
 A. Magos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 206–209 207

therefore comparatively cheap, utilises software for capture

and real-time compression which is freely available, and has
good recording and editing capabilities.

2. Materials and methods

Continuous visual images of endoscopic, vaginal and

laparotomy procedures were obtained using a standard
three-chip endoscopic camera system (Karl Storz, Tut-
tlingen, Germany); in the case of conventional surgery, the
camera was fixed to a modified microphone stand. All
images were displayed on the colour monitor on the
endoscopy stack and routed to the computer system.
Fig. 1. Schematic diagram of the digital recording system.
2.1. Computer hardware

The hardware specification, cost and recording cap-

abilities of our video capture system and that of three
commercially available systems are summarised in
sTable 1. A modern, off-the-shelf desktop or portable
computer has more than enough computing power for
video capture and editing; a desktop PC has the dual
advantages that it can be fitted with a faster and larger
hard disk (resulting in better recording quality and lomger
recording capacity, respectively). Video capture in high-
quality DV format is facilitated by the use of an external
analogue-digital converter which is linked to the computer
via a fast firewire cable (Fig. 1).
2.2. Computer software
The software we utilised for capturing, compressing and
editing movie files are available on the internet and can be
downloaded and installed free of charge; DivXTM is free
provided it is for personal and not commercial use.
The digital signal was captured in real-time using the
video capture application VirtualVCR (available from
http://virtualvcr.sourceforge.net). Usefully, the program
provides real-time capture statistics of variables such as
the number of frames captured, dropped frames, compres-
sion ratio, size of the file and the time left on the hard disk
for further recording. The capture statistics can be saved
along with the movie file if desired. As an added bonus
with VirtualVCR, the teaching of trainees and students is
facilitated as the standard Windows pointer is available on
the colour monitor to demonstrate the anatomy and
highlight steps involved in the surgical procedure. If
desired, screen shots can be captured and saved using
freeware programs such as ScreenPrint32 (available from
http://www.provtech.co.uk/software/screenprint32.asp) or
WinGrab (available from http://home.no.net/wingrab) as
an alternative to post-recording frame capture with
VirtualDub (see below).

Table 1
System specifications and cost
Desktop Laptop Storz Olympus Stryker
computer computer AIDA UPAP100MD SDCPro II
system system compact
Computer (e.g. Intel Pentium 42.8 GHz processor, £ 600 £ 1000a
512 RAM, 250 Gb 7200 rpm hard disk, firewire port
(IEEE-1394), DVD/CD writer, keyboard and mouse,
Windows XP operating system)
External analogue-digital converter £ 220 £ 220
(e.g. Canopus ADVC-100)
Colour monitor £ 100 Included
Video capture software (e.g. VirtualVCR) Free Free Included Included Included
Video compression software (e.g. DivXTM MPEG-4) Free Free Included Included Included
Video editing software (e.g. VirtualDub) Free Free N/A N/A N/A
Colour photoprinter (e.g. HP Photosmart 130) £ 80 £ 80 Included Not included Not included
Maximum continuous recording time 230 h 55 ha 45 min 4 min 60 min
Total recommended retail cost (excluding tax) £ 1000, £ 1300, £ 9999, £ 5250, £ 13000,
s 1430, s 1859, s 14299, s 7507, s 18590,
$ 1720 $ 2236 $ 17198 $ 9030 $ 22360
a
Hard disk size 60 Gb.
208 A. Magos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 206–209
As video images are relatively memory hungry (e.g.
10 min of uncompressed video requires about 15 Gb hard
disk storage), it is advisable to compress the video,
preferably in real-time (that is, as the procedure is being
recorded) to avoid delays with subsequent recordings. We
used software compression based on the DivXTM MPEG-4
encoder/decoder (available from http://ww.divx.com) which
allows good quality compression at lower bitrates (computer
memory required for a second of recording). The properties
of the codec can be altered within VirtualVCR in terms of
the bitrate (compression), frame rate and frame size
(resolution) amongst others, factors which will determine
the file size and quality of the recording; for instance, higher
compression will result in a smaller file size but poorer
quality.
The recording, in AVI format, can be edited to show
highlights of the surgery using VirtualDub (available from
http://www.virtualdub.org). Editing can be done on a frame-
by-frame basis, and individual frames can be easily copied
and pasted into standard photoediting programs (e.g.
Microsoft1 PhotoEditor, Adobe1 PhotoShop) to create
still images. It is also possible to apply numerous freely
available filters when editing with VirtualDub to change the
size of the output, crop edges, adjust the colour balance,
apply logos, etc. Such editing options are either very limited
or not available with commercial systems.
The finished movie can be played using computer
software such as Windows1 Media Player which comes free
with the Windows1 operating system for PCs. In addition,
movies and still pictures can be integrated into presentation
software such as Microsoft1 PowerPoint. Still pictures can
be printed to provide hard copies for patients and medical
notes, and movies can be burnt on to CD or DVD.
3. Results
We have been using the system detailed in the table for 9
months, and have made complete recordings of over 100
hysteroscopic (e.g. diagnostic hysteroscopy, polypectomy,
myomectomy, endometrial ablation), laparoscopic (ovarian
cystectomy, adhesiolysis, ablation/excision of endometrio-
sis, myomectomy, hysterectomy), vaginal (hysterectomy)
and open (myomectomy, hysterectomy) procedures.
Our system was significantly cheaper than commercial
systems and yet considerably more versatile; in particular,
complete operations could be recorded continuously without
concern that the recording capacity would be exceeded
(Table 1).
After experimentation with different compression set-
tings, we ultimately set DivX to encode at a maximum 1-
pass bitrate of up to 3000 kbps at frame rate of 25 frames/s at
a resolution of 640  480. This produced an approximately
11-fold compression (e.g. 10 min of compressed video
required about 180 Mb hard disk storage), and twice the
compression of conventional DVD recordings in MPEG-2
Fig. 2. Hard disk memory requirements for 1 h recording at different
compression standards.
(Fig. 2). In the UK where the video standard is PAL, we
found that these settings provided good quality recording for
teaching, presentations and video projection whilst making
the file size manageable and minimising the number of lost
frames; the settings can be altered to provide higher quality
or greater compression depending on the clinical need. The
compression settings could also be easily adjusted to
optimise the system for other video standards such as NTSC
(in the USA) and Secam (in France).
The system was straightforward to set-up and use, and
even those not used to computers could use the system after a
simple demonstration. The capture of still images during live
recording produced high quality pictures and did not
compromise the quality of the recording. Post-recording
editing of the operation movie was easy and much quicker
than the editing of recordings on VHS tape.
4. Discussion
The ability to record surgical procedures digitally has
numerous advantages over analogue recording using tape-
based video recorders. VHS tapes are bulky, prone to
deterioration with time, and editing is time consuming with a
loss of quality on copying. In contrast, digital recordings can
be stored on slim CD or DVD disks, editing is fast, and
quality is not compromised by editing or copying.
Apart from the ease of integration with presentation
software or for printing stills as photographs, the availability
of digital images has other important practical advantages.
Stills and short segments of movies can be posted
electronically to colleagues for an opinion or as part of
the referral process. They can be published in image
databases on the internet for clinical and educational
applications [5]. Video images are also integral to teaching
endoscopic techniques to trainees, and can be used to
measure performance and assess instrumentation [6–8].
Digital videos can now even be submitted to medical
journals as visual supplements to scientific articles [9].
 A. Magos et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 120 (2005) 206–209
Although the system we have described is not as
convenient to use as commercially available products and
utilises the Windows1 desktop rather than a dedicated
interface, its capacity is infinitely greater at a fraction of the
cost. In particular, the availability of a large hard disk to
store data means that the surgeon does not have to plan prior
to the surgery which parts of the procedure will be recorded.
As most surgical procedures take considerably more than
10–15 min [10] and as it is generally impossible to predict
which part of the surgery will be of interest, it is clearly
better to record the entire operation rather than a short
segment. As the recording time is limited with commercially
available systems, this is impossible unless the procedure
time is relatively short. In contrast, the desk-top computer
system we specified in Table 1 can record and retain more
than 200 h of surgery before the hard disk became full. Then,
the movies could be transferred to an external hard disk if
complete recordings were to be kept or edited versions
retained to create space for future procedures. Just as any
computer, the PC could be used for collecting and analysing
research and audit data in addition to such recordings.
We chose software rather than hardware-based capture
and compressing for reasons of cost but also because, unlike
with the latter, all the three programmes, VirtualVCR,
DivXTM and VirtualDub, are ‘‘future proof’’ and can be
updated for free. Similarly, we decided to use free software
for capturing, compressing and editing the digital signals
rather than commercial programs such as Adobe1
Premiere1, Ulead1 VideoStudioTM or Pinnacle Studio
partly because of cost, but also because of the convenience
of real-time compression; whereas the VirtualVCR/DivXTM
combination captures and compresses in real-time, the
commercial programmes process and compress the video
signal only after it has been captured, a process which can
take a considerable time with longer operations with the
result that any subsequent recordings have to be delayed.
This also applies to Windows1 Movie Maker which is
available as a free download from Microsoft1. As a further
advantage for the free programmes, they are also relatively
easy to use with extensive help being available on the
internet should there by any specific problems.
Real-time video capture and compression is also possible
using dedicated graphics and TV cards. However, at present
these cards can only compress using MPEG-1 and MPEG-2
codecs. Although MPEG-1 compression results in small file
size, quality is low and the image size is small; in contrast,
MPEG-2 compression results in DVD quality and a large
image size, but at the expense of large file size (e.g. >2 Gb/
h). Encoding to MPEG-4 produces the best of both worlds,
namely a relatively large image of good quality at a
comparatively small file size.
View publication stats
209
In summary, the computer based digital recording system
we have described is relatively cheap, versatile and has a
huge capacity to record surgical procedures. It is an
excellent teaching aid and has several features which are not
available on current commercial systems. Its use is
applicable to a wide range of specialities performing
endoscopic surgery, and with appropriate patient consent,
could also be used for any type of surgery which would
benefit from being recorded.
5. Addendum
It should be noted that local guidelines must be followed
when any visual recordings are made of patients. In the UK,
for instance, the General Medical Council has issued
guidelines which state that, in the case of laparoscopic
images or images of internal organs, permission or consent is
not required from patients provided the recording are
effectively anonymised by removal of any identifying marks
(further details are available at http://www.gmc-uk.org/
standards/AUD_VID.HTM).
6. Condensation
High quality digital recording of surgical procedures is
possible using a standard personal computer.
References
[1] Darzi A, Mackay S. Recent advances in minimal access surgery. Br
Med J 2002;324:31–4.
[2] Soper NJ, Brunt LM, Kerbl K. Laparoscopic general surgery. N Eng J
Med 1994;330:409–19.
[3] Mack MJ. Minimally invasive and robotic surgery. JAMA
2001;285:568–72.
[4] Rattner DW. New technologies: future directions in innovative mini-
mally invasive surgery. Lancet 1999;353(Suppl. 1):1215.
[5] Kimura N, Nakajima I, Juzoji H, et al. Video endoscopic database on
WWW linking with ISDN. J Med Syst 2001;25:1–7.
[6] Royston CMS, Lansdown MRJ, Brough WA. Teaching laparoscopic
surgery: the need for guidelines. Br Med J 1994;308:1023–5.
[7] Ward M, Macrae H, Schlachta C, et al. Resident self-assessment of
operative performance. Am J Surg 2003;185:521–4.
[8] Heijinskijk EA, Dankelman J, Gouma DJ. Effectiveness of grasping
and duration of clamping using laparoscopic graspers. Surg Endosc
2002;16:1329–31.
[9] Wechsler AS, Fried PW. Editorial: video comes to the journal. J
Thoracic Cardiovasc Surg 2002;124:884–5.
[10] Shushan A, Mohamed H, Magos A. How long does laparoscopic
surgery really take? Lessons learned from 1000 operative laparosco-
pies Hum Reprod 1999;14:39–43.