You are on page 1of 7

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS.

For personal use only; all rights reserved

Computer-Aided Detection (CAD)


in Screening Mammography:
Sensitivity of Commercial CAD Systems
for Detecting Architectural Distortion
Jay A. Baker1
Eric L. Rosen
Joseph Y. Lo
Edgardo I. Gimenez
Ruth Walsh
Mary Scott Soo

OBJECTIVE. Computer-aided detection (CAD) algorithms have successfully revealed


breast masses and microcalcifications on screening mammography. The purpose of our study
was to evaluate the sensitivity of commercially available CAD systems for revealing architectural distortion, the third most common appearance of breast cancer.
MATERIALS AND METHODS. Two commercially available CAD systems were used
to evaluate screening mammograms obtained in 43 patients with 45 mammographically detected regions of architectural distortion. For each CAD system, we determined the sensitivity
for revealing architectural distortion on at least one image of the two-view mammographic
examination (case sensitivity) and for each individual mammogram (image sensitivity). Surgical biopsy results were available for each case of architectural distortion.
RESULTS. Architectural distortion was deemed present and actionable by a panel of expert
breast imagers in 80 views of the 45 cases. One CAD system detected distortion in 22 of 45 cases
of distortion (case sensitivity, 49%) and in 30 of 80 mammograms (image sensitivity, 38%); it
displayed 0.7 false-positive marks per image. Another CAD system identified distortion in 15 of
45 cases (case sensitivity, 33%) and 17 of 80 mammograms (image sensitivity, 21%); it displayed
1.27 false-positive marks per image. Sensitivity for malignancy-caused distortion was similar to
or lower than sensitivity for all causes of distortion.
CONCLUSION. Fewer than one half of the cases of architectural distortion were detected by the two most widely available CAD systems used for interpretations of screening
mammograms. Considerable improvement in the sensitivity of CAD systems is needed for
detecting this type of lesion. Practicing breast imagers who use CAD systems should remain
vigilant for architectural distortion.

Received December 23, 2002; accepted after revision


April 22, 2003.
Presented at the annual meeting of the American
Roentgen Ray Society, San Diego, May 2003.
1
All authors: Department of Radiology, Duke University
Medical Center, Box 3808, Durham, NC 27710.
Address correspondence to J. A. Baker
(jay.baker@duke.edu).

AJR 2003;181:10831088
0361803X/03/18141083
American Roentgen Ray Society

AJR:181, October 2003

ecent studies have proven that


computer-aided detection (CAD)
algorithms are capable of revealing breast lesions on screening mammography
and of reducing the number of false-negative
mammographic findings [14]. Studies typically report the sensitivity of commercially
available devices and novel algorithms for microcalcifications and breast masses separately.
The sensitivity of CAD systems for the detection of malignant microcalcifications has been
reported to be as high as 99% [3], whereas the
sensitivity of the systems for the detection of
malignant breast masses has been reported to
be 7589% [3, 5, 6] in large series.
Although most breast cancers are identified
on screening mammography as either a breast
mass or a focus of microcalcifications, the
third most common mammographic appearance of nonpalpable breast cancer is architec-

tural distortion (i.e., a distortion of the


parenchymal architecture without a concomitant mass) [79]. Because architectural distortion may mimic the normal appearance of
overlapping breast tissue, this finding can be
subtle and may be particularly difficult to detect [10]. Architectural distortion is a worrisome finding and has been reported to
represent a breast malignancy in from one half
to two thirds of the cases in which it is present
[8, 9]. Because of its subtlety and potential for
malignancy, architectural distortion is a common cause of false-negative findings on
screening mammograms [1113]. Therefore,
evaluating the sensitivity of CAD systems for
revealing this subtle appearance of breast cancer is valuable. We sought to determine the
sensitivity of two commercially available
CAD systems for identifying architectural distortion on routine screening mammograms.

1083

Baker et al.
Materials and Methods

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

Case Collection and Study Population


We obtained institutional review board approval
for our study; informed consent was not required.
Using a computerized database of patients who had
undergone biopsy at our institution, we identified
51 cases of architectural distortion and no associated findings in 49 women (age range, 3382
years) who had been imaged between August 1996
and October 2001. Each of these cases of architectural distortion had been prospectively described at
the time of the clinical study as showing only architectural distortion with no associated mass, microcalcifications, or other findings. Final histologic
analysis after wire localization and surgical excision was available for all cases.
Although each of the 51 cases was prospectively
reported as architectural distortion, the rate of interobserver variability in choosing breast lesion descriptors [14] is high. Therefore, we had a panel of five
breast radiologists, each of whom had at least 4 years
of experience, determine the final morphology descriptor for each lesion in our study. These radiologists evaluated the routine craniocaudal and
mediolateral oblique mammograms for the 51 cases.
The panel members were unaware of the results of
any additional imaging or the biopsy. For our study, a
lesion was judged to be an architectural distortion if
most (i.e., three of five) of the reviewers identified it
as such on a particular mammogram. Each reviewer
independently determined whether each lesion was
best described as an architectural distortion without
associated findings, a mass or focal density, or a nonactionable lesion because the lesion was not sufficiently conspicuous. This assessment was made for
both the craniocaudal and mediolateral projections.
Of the 51 lesions described prospectively at the
time of biopsy as architectural distortion, six were excluded because most of our panelists labeled the lesion as either a spiculated mass (five lesions) or a
focal density (one lesion). The panel judged 45 lesions in 43 patients to be architectural distortions
without associated findings; these 45 lesions composed our study population. One patient presented
with bilateral synchronous foci of architectural distortion, and one patient presented with two foci of synchronous architectural distortion in the same breast.
CAD Analysis
Routine screening mammograms (craniocaudal
and mediolateral) from each patient were analyzed
using two commercially available CAD systems:
ImageChecker M1000, version 2.5 (R2 Technology, Sunnyvale, CA) and SecondLook, version 4.0
(CADx Medical Systems, Laval, QC, Canada). To
create a digital image of each mammographic view,
we used the standard digitizer included with each
of the CAD systems (50-mm resolution for the R2
ImageChecker; 43.5-mm resolution for the CADx
SecondLook). The digital images were then analyzed using proprietary software (included with the
CAD systems) designed to identify breast cancers
presenting as microcalcifications or mass lesions.

1084

Such systems identify masses by searching for a


central density with radiating lines, suggesting spiculation, or for radiating lines without a central density, suggesting architectural distortion [4, 6, 15].
The output of the CAD system can be displayed on
either two small video monitors or a single large flatpanel monitor or can be printed to paper, depending
on the system and configuration. The CAD output
shows suspicious foci of calcification and masses
marked on a low-resolution reproduction of the
mammograms. The R2 system uses small triangles
to mark the location of possible calcifications and
asterisks to mark possible masses; the CADx system
uses rectangles to mark possible calcifications and
ovals to mark possible masses.
Case Evaluation
The precise location of the architectural distortion was determined on both the craniocaudal and
mediolateral mammograms through consensus by
two breast radiologists with fellowship subspecialty
training in breast imaging and at least 4 years experience. These radiologists had access to all diagnostic images and wire localization images so that they
could determine the actual location of the lesion on
the craniocaudal and mediolateral oblique mammograms. They then compared this location with the location marked by the CAD systems to determine
whether the distortion on each mammogram was
correctly identified by each CAD system.
No universally accepted rule exists to determine
whether a particular CAD mark is sufficiently close
to a lesion to represent a true-positive mark [16].
Therefore, the two reviewing radiologists reached
the decision of whether each lesion was correctly
marked through consensus. When evaluating cases
analyzed by the R2 ImageCheckerwhich places
an asterisk at the site of a possible masswe judged
a CAD mark to be a true-positive if the asterisk was
anywhere within the boundaries determined by the
two radiologists as outlining the architectural distortion. In an effort to be fair, when evaluating CADx
SecondLookwhich places a variably sized oval to
encompass the location of a possible masswe
judged an oval to be a true-positive mark if the center of the oval was anywhere within the boundaries
outlining the distortion. Under a more lenient system, we might have allowed a portionother than
the centerof the oval that overlapped part of the
architectural distortion to be counted as a true-positive mark; however, we did not encounter such a circumstance in any of the 45 lesions in our study.
The sensitivity of each CAD system for identifying architectural distortion in the 43 patients with 45
foci of architectural distortion was determined. The
case sensitivity was determined by dividing the
number of cases in which architectural distortion
was correctly marked on either the craniocaudal or
mediolateral mammogram by the total number of
cases of architectural distortion. The image sensitivity was determined by dividing the number of mammograms on which architectural distortion was
marked by the total number of projections on which

it was visible (i.e., one or two mammograms per


case). These sensitivities were calculated for both of
the commercially available CAD systems we studied. In addition, the case sensitivity and image sensitivity for malignant lesions were determined for
each CAD system. The McNemar test was used to
compare the sensitivities of the two CAD systems.
Because there may be one or more false-positive
marks per patient, specificity in CAD studies has
historically been determined by the number of
false-positive marks per image or per patient. For
the purposes of our study, only marks that correctly
indicated the location of architectural distortion
were considered true-positive marks. Malignant lesions presenting as masses with appearances other
than distortion or presenting as calcifications were
excluded. The average number of false-positive
marks per mammogram was determined for each
CAD system, and statistical comparison of falsepositive marks for the two systems was performed
using the Students paired t test.

Results

Of the 45 cases of architectural distortion,


35 were judged by most of the five-radiologist panel to have architectural distortion
visible on both craniocaudal and mediolateral mammograms. The remaining 10 cases
were judged to have the distortion visible in
only one of the two projections. Therefore,
architectural distortion was reported by most
of the panel to be present on 80 separate
mammograms for the 45 cases of architectural distortion.
In all patients, the lesions were surgically
excised. At the histopathologic examination,
the focus of architectural distortion was
found to be malignant in 27 (60%) of the 45
cases. Ten (22%) of the 45 cases represented
invasive ductal carcinoma (Fig. 1), 10 (22%)
cases represented invasive ductal carcinoma
with ductal carcinoma in situ (DCIS), and
two cases (4%) represented invasive lobular
carcinoma. DCIS was the sole finding in five
(11%) of the 45 cases (Fig. 2). The radiologist panel determined that architectural distortion was visible (and therefore actionable)
on 51 images of the 27 malignant lesions.
The lesion could not be identified on one of
the two-view mammograms in three cases of
malignancy, and the panel therefore deemed
the lesion not actionable on that mammographic view.
Eighteen (40%) of the 45 cases of architectural distortion were benign. Twelve
(27%) of the 45 cases represented a radial
scar or a complex sclerosing lesion. The remaining six benign lesions (13%) included
four cases of intralobular fibrosis, one case

AJR:181, October 2003

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

CAD in Screening Mammography

Fig. 1.59-year-old woman with invasive ductal carcinoma of right breast. Architectural distortion seen on images was successfully detected and marked by only one of
two computer-aided detection systems tested.
A, Craniocaudal mammogram shows typical appearance of architectural distortion (box): radiating lines without central density.
B, Mediolateral oblique mammogram also shows architectural distortion (box).
C, Spot compression magnification image of right breast shows lack of central density at site of distortion (arrow) more clearly than do mammograms A and B.

Fig. 2.55-year-old woman with ductal carcinoma in situ of left breast. Architectural distortion was successfully identified by interpreting radiologist but was not detected
by either computer-aided detection system.
A, Craniocaudal mammogram shows subtle architectural distortion (box) in lateral aspect of breast. Note radiating lines without central mass in dense breast parenchyma.
B, Mediolateral oblique mammogram shows more conspicuous architectural distortion (box) than craniocaudal view (A). No central density is present.
C, Specimen radiograph from wire-localized surgical excision confirms architectural distortion (arrow) centered on middle of thickened wire. No mass or associated calcifications are seen.

AJR:181, October 2003

1085

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

Baker et al.
of benign proliferative change, and one case of
a surgically confirmed postoperative scar.
The R2 ImageChecker system correctly
marked at least one of the two screening mammographic views as containing a possible
mass at the correct location in 22 of the 45
cases of architectural distortion (case sensitivity, 49%). The CADx SecondLook system correctly marked 15 cases of architectural
distortion (case sensitivity, 33%). We found a
trend toward better case sensitivity for the ImageChecker system for detection of both benign and malignant causes of architectural
distortion, but the difference between the ImageChecker and the SecondLook systems did
not reach the level of statistical significance
( p = 0.10).
The ImageChecker system had virtually
identical sensitivity for detecting malignant
causes of architectural distortion as it did for
detecting all cases of distortion. In malignant
cases, the ImageChecker system successfully
identified 13 (48%) of 27 cases of malignant
distortion. The rate of detection for malignant
cases for the ImageChecker system was significantly higher than that for the SecondLook
system, which identified only five (19%) of the
27 malignant lesions ( p = 0.027).
Architectural distortion without associated
findings was seen on 80 mammographic
views35 cases in which it was visible in
both craniocaudal and mediolateral views and
10 cases in which it was visible in only one of
the two views. The focus of distortion was correctly identified in 30 of the 80 views (image
sensitivity, 38%) by the ImageChecker system,
significantly better than the 17 of 80 views correctly marked by the SecondLook system (image sensitivity, 21%) ( p = 0.01).
The ImageChecker system was also significantly more successful at detecting malignant
foci of distortion on each image (view) in which
it was deemed actionable by the panel of radiologists. This system successfully detected the
malignancy in 16 of the 51 images in which the
distortion represented breast cancer (image sensitivity, 31%) compared with five of the 51 images (image sensitivity, 10%) for the
SecondLook system (p = 0.01).
The two CAD systems marked different
subsets of lesions (Fig. 1) as possible malignancies. Of the 45 cases, nine cases (20%)
were successfully identified on at least one
view by both CAD systems. The ImageChecker system identified 13 cases (29%)
of architectural distortion that were not identified by the SecondLook system. In comparison, the SecondLook system identified five

1086

cases (11%) that were not identified by the ImageChecker system. Eighteen (40%) of the 45
cases of architectural distortion were not detected by either CAD system (Fig. 2).
Each CAD system also displayed a number
of false-positive marks. On average, the ImageChecker system displayed 0.70 false-positive marks per image. This rate was
statistically less than the 1.27 false-positive
marks per image displayed by the SecondLook
system ( p < 0.0001).

Discussion

CAD systems for mammography have been


under development for at least 35 years, with
the first algorithm reported in 1967 by Winsberg
et al. [17]. Subsequent studies have described
numerous detection algorithms and promising
clinical outcomes [14, 15] for computer assistance in identifying breast lesions. A recent
study of CAD interpretations of screening
mammograms found an overall improvement of
almost 20% in the CAD detection of breast cancer; microcalcifications accounted for most of
the malignancies identified only by the CAD
system [2]. However, previous CAD studies
have focused either on microcalcifications or
breast masses or both. To our knowledge, no
studies have focused specifically on the detection of architectural distortion.
Architectural distortion has been described
by the American College of Radiology in its
Breast Imaging Reporting and Data System
[18] as [t]he normal architecture is distorted
with no definite mass visible. This includes
spiculations radiating from a point, and focal
retraction or distortion of the edge of the parenchyma. The differential diagnosis of architectural distortion includes malignant lesions
such as invasive ductal carcinoma, invasive
lobular carcinoma, and DCIS and benign lesions such as a surgical scar, a radial scar, a
complex sclerosing lesion, fat necrosis, and intralobular fibrosis [10]. Most malignancies
presenting as architectural distortion are invasive rather than in situ carcinomas. In one
study of breast malignancies, two thirds (16/
24) of the cases presenting as architectural distortion were invasive ductal carcinoma, 21%
(5/24) were invasive lobular carcinoma, and
13% (3/24) were DCIS alone [19]. These figures parallel our study results, with invasive
ductal and lobular carcinomas accounting for
81% (22/27) of the malignancies.
Although several benign entities are included
in the differential diagnosis, the malignancy
rates for architectural distortion range from al-

most one half to two thirds of the cases [8, 9].


Unfortunately, mammography cannot be used
to differentiate benign from malignant foci of
architectural distortions. Several studies have
documented that mammographic features such
as the length of radiating lines or the presence of
a central density cannot be used to differentiate
benign from malignant lesions [10, 2022]. Because architectural distortion is frequently
caused by malignancy and because benign
causes of distortion cannot be excluded on the
basis of imaging features, identification of all
cases of distortion is essential, as is performance
of a definitive biopsy.
Architectural distortion has been reported to
be the third most frequent mammographic appearance of breast cancer [7, 8], but distortion
may be challenging to detect. In one study of
malignancies overlooked by radiologists on
screening mammograms, architectural distortion accounted for nine (12%) of the 77 missed
cases [13]. Another study placed the percentage of missed malignancies presenting as architectural distortion as high as 45% [11].
Despite the subtlety and potential for malignancy of architectural distortion, few reports
investigating the efficacy of CAD algorithms
have specifically addressed detection of distortion. In one study, Evans et al. [4] investigated
CAD sensitivity for the detection of lobular
carcinoma and found that 17 (85%) of 20
cases of lobular carcinoma presenting as architectural distortion were successfully marked
by a CAD system. Likewise, Birdwell et al. [1]
evaluated 115 breast cancers overlooked by
the interpreting radiologist and found that five
(83%) of six missed breast cancers presenting
as architectural distortion were successfully
identified by one CAD system.
However, our results indicated that the two
most widely available commercial CAD systems had only limited success in detecting architectural distortion. The more successful
system in our study identified distortion on at
least one mammographic view in slightly
fewer than half of our cases, whereas the other
system identified just one third of cases. Compared with their success in identifying calcifications and masses, the success of these
systems was substantially less in identifying
architectural distortion. Of the 80 mammographic views on which a distortion was
deemed visible and actionable by most members of the radiologist panel, only one case in
five (on average) was identified by one of the
CAD systems, with the more successful system performing only moderately better than
the less successful system. Particularly trou-

AJR:181, October 2003

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

CAD in Screening Mammography


bling is the finding that the sensitivity of both
systems was either the same or moderately
worse in detecting distortions caused by malignancy as in detecting distortions due to benign causes such as radial scars.
Our study is limited in that it only examined
the sensitivity of CAD systems for those foci
of architectural distortion initially detected by
a human observer. Theoretically, a CAD system can detect other foci of architectural distortion that may have been overlooked by a
human observer until prompted to further evaluate the area by a CAD mark. Therefore, a
CAD system could actually detect a higher
percentage of architectural distortion lesions
than we found in our study if the cases that
were only identified by a CAD system were
included. Future research on the sensitivity of
CAD systems to specific mammographic features could include a large trial with prospective descriptions of the morphology of all
lesions detected by a human observer or a
CAD system.
Although CAD systems are largely successful at identifying breast masses and microcalcifications, the finding that they are less
successful at identifying architectural distortion is not surprising, given the techniques
used in the CAD analysis of mammograms.
Many algorithms for breast mass detection
rely heavily on the presence of a central density [4, 5, 2330]. Techniques such as template
matching [24] and low-pass or band-pass filtering (e.g., gaussian filtering) [30, 31] function by searching for a region that is relatively
more dense than the surrounding tissue; leftto-right subtractions search for asymmetric
density [27, 28]. Such algorithms are not designed to recognize the radiating lines that define architectural distortion.
Other experimental CAD algorithms may
identify radiating lines. Examples include radial-edge gradient-based algorithms [32, 33],
edge profile acuteness (i.e., sharpness) measurements [34], and rubber band straightening
transform analysis [35] for evaluating mass
edge features. Such techniques have the potential to be more sensitive in detecting architectural distortion, regardless of the presence of a
central density.
Because the techniques used by commercial
vendors are proprietary, we cannot determine
which mass-detection algorithms each system
uses. Prior reports have indicated that the ImageChecker M1000 software searches for features common to malignant masses such as areas
with central density and radiating lines [4]. The
fact that the CAD system used by Evans et al.

AJR:181, October 2003

[4] had lower sensitivity for detecting architectural distortion than for detecting other lesions
may be explained in part by the researchers
statement that [w]hen no central density is
found, the radiating lines must be more pronounced to be marked. This approach limits the
number of false-positive marks for each case because normal overlapping tissue (e.g., Coopers
ligaments) can mimic the radiating lines of architectural distortion, deceiving both a radiologist and a CAD algorithm. Each of the other
commercial systems approved by the United
States Food and Drug Administration, including
SecondLook and the more recently approved
MammoReader (iCAD, Nashua, NH), must
make similar trade-offs between sensitivity and
false-positive marks.
In our study, the more sensitive of the two
systems (ImageChecker) also had a significantly lower number of false-positive marks
per image. Systems that generate many falsepositive marks may result in a true-positive
mark being ignored by a radiologist overwhelmed by distracting prompts. Therefore,
the false-positive rate of a CAD system must
be considered along with its sensitivity.
The purpose of our investigation was to test
the sensitivity of increasingly available CAD
systems to determine whether such systems
are as successful in detecting worrisome foci
of architectural distortion as they are in detecting more common breast masses and calcifications. Although one of the CAD systems was
significantly more sensitive than the other for
detecting architectural distortion, a study by
Nelson et al. [36] found that the three commercially available mammography CAD systemsthe R2 ImageChecker, the CADx
SecondLook, and the iCAD MammoReader
all performed with nearly identical sensitivity
in a study of 128 malignant masses and clusters of calcifications. We found that both systems had substantially lower rates for
identifying architectural distortion than the
previously reported rates of those systems for
detecting more common masses and calcifications. Clearly, both systems need to be improved, given that one half to two thirds of the
cases of architectural distortion were not identified by the two most widely available commercial CAD systems.
Because of the similarity between architectural distortion and overlapping fibroglandular
tissue, improvement in detection may prove
difficult without a concomitantand perhaps
unacceptableincrease in the number of
false-positive marks per image. Nevertheless,
now that CAD systems can successfully iden-

tify almost all malignant calcifications and


most malignant masses, the capability of the
systems to detect the more subtle signs of malignancy, such as architectural distortion,
should be addressed. In fact, despite the introduction of computer systems to assist radiologists in the challenging task of identifying
breast cancers, little has changed since Sickles landmark study of 300 consecutive cases
of nonpalpable breast cancer [7] in which he
concluded that [t]o take advantage of mammography, [radiologists] must search diligently not only for characteristic tumor masses
and clustered calcifications, but especially for
more subtle signs of malignancy (e.g., architectural distortion). Our results have shown
that further work is necessary to develop CAD
systems that are more capable of assisting radiologists in that diligent search.
References
1. Birdwell R, Ikeda D, OShaughnessy K, Sickles
E. Mammographic characteristics of 115 missed
cancers later detected with screening mammography and the potential utility of computer-aided
detection. Radiology 2001;219:192202
2. Freer T, Ulissey M. Screening mammography
with computer-aided detection: prospective study
of 12,860 patients in a community breast center.
Radiology 2001;220:781786
3. Burhenne LW, Wood S, DOrsi C, et al. Potential
contribution of computer-aided detection to the
sensitivity of screening mammography. Radiology 2000;215:554562
4. Evans WP, Burhenne LJ, Laurie L, OShaughnessy
K, Castellino R. Invasive lobular carcinoma of the
breast: mammographic characteristics and computeraided detection. Radiology 2002;225:182189
5. Petrick N, Sahiner B, Chan H, Helvie M, Paquerault
S, Hadjiiski L. Breast cancer detection: evaluation
of a mass-detection algorithm for computer-aided
diagnosis: experience in 263 patients. Radiology
2002;224:217224
6. Malich A, Marx C, Facius M, Boehm T, Fleck M,
Kaiser W. Tumour detection rate of a new commercially available computer-aided detection system. Eur Radiol 2001;11:24542459
7. Sickles EA. Mammographic features of 300 consecutive nonpalpable breast cancers. AJR 1986;
146:661663
8. Knutzen AM, Gisvold JJ. Likelihood of malignant disease for various categories of mammographically detected, nonpalpable breast lesions.
Mayo Clin Proc 1993;68:454460
9. Johnston J, Clee C. Analysis of 308 localisation
breast biopsies in a New Zealand hospital. Australas Radiol 1991;35:148151
10. Kopans D. Breast imaging, 2nd ed. Philadelphia:
Lippincott-Raven, 1998
11. Burrell H, Sibbering D, Wilson A, et al. Screening interval breast cancers: mammograhic features of prognostic factors. Radiology 1996;199:
811817

1087

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

Baker et al.
12. Burrell H, Evans A, Wilson A, Pinder S. Falsenegative breast screening assessment: what lessons can we learn? Clin Radiol 2001;56:385388
13. Bird RE, Wallace TW, Yankaskas BC. Analysis of
cancers missed at screening mammography. Radiology 1992;184:613617
14. Baker JA, Kornguth PJ, Floyd CE Jr. Breast imaging reporting and data system standardized
mammography lexicon: observer variability in lesion description. AJR 1996;166:773778
15. Vyborny C, Doi T, OShaughnessy K, Romsdahl
H, Schneider A, Stein A. Breast cancer: importance of spiculation in computer-aided detection.
Radiology 2000;215:703707
16. Kallergi M, Carney GM, Gaviria J. Evaluating the
performance of detection algorithms in digital
mammography. Med Phys 1999;26:267275
17. Winsberg F, Elkin M, Macy J, Bordaz V, Weymouth
W. Detection of radiographic abnormalities in mammograms by means of optical scanning and computer analysis. Radiology 1967;89:211215
18. American College of Radiology. Breast imaging reporting and data system (BI-RADS), 3rd ed. Reston,
VA: American College of Radiology, 1998
19. Gajdos C, Tartter P, Bleiweiss I, et al. Mammographic appearance of nonpalpable breast cancer
reflects pathologic characteristics. Ann Surg
2002;235:246251
20. Ciatto S, Morrone D, Catarzi S, et al. Radial scars
of the breast: review of 38 consecutive mammographic diagnoses. Radiology 1993;187:757760
21. Mitnick J, Vazquez M, Harris M, Roses D. Differ-

1088

22.

23.

24.

25.

26.

27.

28.

29.

entiation of radial scar from scirrhous carcinoma


of the breast: mammographic-pathologic correlation. Radiology 1989;173:697700
Vega A, Garijo F. Radial scar and tubular carcinoma: mammographic and sonographic findings.
Acta Radiol 1993;34:4347
Petrick N, Chan H, Sahiner B, Wei D. An adaptive density-weighted contrast enhancement filter
for mammographic breast mass detection. IEEE
Trans Med Imaging 1996;15:5967
Hatanaka Y, Hara T, Fujita H, Kasai S, Endo T,
Iwase T. Development of an automated method
for detecting mammographic masses with a partial loss of region. IEEE Trans Med Imaging
2001;20:12091214
Yin FF, Giger ML, Doi K, Vyborny CJ, Schmidt
RA. Computerized detection of masses in digital
mammograms: investigation of feature-analysis
techniques. J Digit Imaging 1994;7:1826
Vyborny CJ, Giger ML. Computer vision and artificial intelligence in mammography. AJR 1994;
162:699708
Yin FF, Giger ML, Vyborny CJ, Doi K, Schmidt
RA. Comparison of bilateral-subtraction and single-image processing techniques in the computerized detection of mammographic masses. Invest
Radiol 1993;28:473481
Mendez AJ, Tahoces PG, Lado MJ, Souto M,
Vidal JJ. Computer-aided diagnosis: automatic
detection of malignant masses in digitized mammograms. Med Phys 1998;25:957964
Mudigonda N, Rangayyan R, Desautels J. Detec-

30.

31.

32.

33.

34.

35.

36.

tion of breast masses in mammograms by density


slicing and texture flow-field analysis. IEEE
Trans Med Imaging 2001;20:12151227
Chang YH, Zheng B, Gur D. Computerized identification of suspicious regions for masses in digitized
mammograms. Invest Radiol 1996;31:146153
Zheng B, Chang YH, Gur D. Computerized detection of masses in digitized mammograms using single-image segmentation and a multilayer
topographic feature analysis. Acad Radiol 1995;
2:959966
Huo Z, Giger ML, Vyborny CJ, et al. Analysis of
spiculation in the computerized classification of
mammographic masses. Med Phys 1995;22:
15691579
Kegelmeyer WP, Pruneda JM, Bourland PD, Hillis A, Riggs MW, Nipper ML. Computer-aided
mammographic screening for spiculated lesions.
Radiology 1994;191:331337
Rangayyan R, El-Faramawy N, Desautels J, Alim
O. Measures of acutance and shape for classification of breast tumors. IEEE Trans Med Imaging
1997;16:799810
Sahiner B, Chan H, Petrick N, Helvie M, Goodsitt M. Computerized characterization of masses
on mammograms: the rubber band straightening
transform and texture analysis. Med Phys 1998;
25:516526
Nelson MT, Lechner MC. Comparison of three
commercially available FDA approved computeraided detection (CAD) systems. (abstr) Radiology
2002;225(P):600

AJR:181, October 2003

Downloaded from www.ajronline.org by 190.179.191.48 on 07/07/13 from IP address 190.179.191.48. Copyright ARRS. For personal use only; all rights reserved

This article has been cited by:


1. Rangaraj M. Rangayyan, Shantanu Banik, Jayasree Chakraborty, Sudipta Mukhopadhyay, J. E. Leo Desautels. 2013. Measures
of divergence of oriented patterns for the detection of architectural distortion in prior mammograms. International Journal of
Computer Assisted Radiology and Surgery 8:4, 527-545. [CrossRef]
2. Xavier Bargall, Martn Velasco, Gorane Santamara, Montse Amo, Pedro Arguis, Sonia Snchez Gmez. 2013. Role of ComputerAided Detection in Very Small Screening Detected Invasive Breast Cancers. Journal of Digital Imaging 26:3, 572-577. [CrossRef]
3. Jennifer A. Harvey, David E. MarchArchitectural Distortion 240-265. [CrossRef]
4. Dianne Georgian-Smith, Richard H. Moore, Elkan Halpern, Eren D. Yeh, Elizabeth A. Rafferty, Helen Anne D'Alessandro, Mary
Staffa, Deborah A. Hall, Kathleen A. McCarthy, Daniel B. Kopans. 2007. Blinded Comparison of Computer-Aided Detection
with Human Second Reading in Screening Mammography. American Journal of Roentgenology 189:5, 1135-1141. [Abstract] [Full
Text] [PDF] [PDF Plus]
5. Per Skaane, Ashwini Kshirsagar, Sandra Stapleton, Kari Young, Ronald A. Castellino. 2007. Effect of Computer-Aided Detection
on Independent Double Reading of Paired Screen-Film and Full-Field Digital Screening Mammograms. American Journal of
Roentgenology 188:2, 377-384. [Abstract] [Full Text] [PDF] [PDF Plus]
6. Justin M. Ko, Michael J. Nicholas, Jeffrey B. Mendel, Priscilla J. Slanetz. 2006. Prospective Assessment of Computer-Aided
Detection in Interpretation of Screening Mammography. American Journal of Roentgenology 187:6, 1483-1491. [Abstract] [Full
Text] [PDF] [PDF Plus]
7. Stephen H. Taplin, Carolyn M. Rutter, Constance D. Lehman. 2006. Testing the Effect of Computer-Assisted Detection on
Interpretive Performance in Screening Mammography. American Journal of Roentgenology 187:6, 1475-1482. [Abstract] [Full Text]
[PDF] [PDF Plus]
8. M. Camille Alexander, Bonnie C. Yankaskas, Karl W. Biesemier. 2006. Association of Stellate Mammographic Pattern with
Survival in Small Invasive Breast Tumors. American Journal of Roentgenology 187:1, 29-37. [Abstract] [Full Text] [PDF] [PDF
Plus]
9. R. James Brenner, Michael J. Ulissey, Ronald M. Wilt. 2006. Computer-Aided Detection as Evidence in the Courtroom: Potential
Implications of an Appellate Court's Ruling. American Journal of Roentgenology 186:1, 48-51. [Abstract] [Full Text] [PDF] [PDF
Plus]
10. Mary Scott Soo, Eric L. Rosen, Jessie Q. Xia, Sujata Ghate, Jay A. Baker. 2005. Computer-Aided Detection of Amorphous
Calcifications. American Journal of Roentgenology 184:3, 887-892. [Abstract] [Full Text] [PDF] [PDF Plus]
11. References 20041435, 1187-1261. [CrossRef]