Computer-aided diagnosis (CAD) methods are being developed to help radiologists improve the interpretation of mammograms for the detection of breast cancer. We review several laboratory observer performance studies of computer-aided diagnosis of malignant and benign breast lesions. These studies show that CAD can improve radiologists' diagnostic performance by increasing the number of their biopsy recommendations for actual malignant lesions while reducing the number of their biopsy recommendations for suspicious but actually benign lesions, and by reducing the variability in their interpretation of mammograms. These results indicate a potential clinical role of CAD in mammography for the detection of breast cancer.
References
1.
JemalA., ThomasA., MurrayT., ThunM., Cancer statistics, 2002. CA Cancer J Clin52, 23–47 (2002).
2.
ElmoreJ. G., BartonM. B., MoceriV. M., PolkS., ArenaP. J., FletcherS. W., Ten-year risk of false positive screening mammograms and clinical breast examinations. N Engl J Med338, 1089–1096 (1998).
3.
TabárL., DeanP. B., Teaching Atlas of Mammography. Thieme-Stratton, New York, (1985).
4.
American College of Radiology. Breast Imaging Reporting and Data System (BI-RADS™). Third ed.American College of Radiology, Reston, VA, (1998).
5.
JiangY., NishikawaR. M., WolvertonD. E., MetzC. E., GigerM. L., SchmidtR. A., VybornyC. J., DoiK., Malignant and benign clustered microcalcifications: Automated feature analysis and classification. Radiology198, 671–678 (1996).
6.
GettyD. J., PickettR. M., D'OrsiC. J., SwetsJ. A., Enhanced interpretation of diagnostic images. Invest Radiol23, 240–252 (1988).
7.
ChanH. P., SahinerB., HelvieM. A., PetrickN., RoubidouxM. A., WilsonT. E., AdlerD. D., ParamagulC., NewmanJ. S., Sanjay-GopalS., Improvement of radiologists' characterization of mammographic masses by using computer-aided diagnosis: An ROC study. Radiology212, 817–827 (1999).
8.
JiangY., NishikawaR. M., SchmidtR. A., MetzC. E., GigerM. L., DoiK., Improving breast cancer diagnosis with computer-aided diagnosis. Acad Radiol6, 22–33 (1999).
9.
SwetsJ. A., PickettR. M., Evaluation of Diagnostic Systems: Methods from Signal Detection Theory. Academic Press, New York, (1982).
10.
MetzC. E., Some practical issues of experimental design and data analysis in radiological ROC studies. Invest Radiol24, 234–245 (1989).
11.
MetzC. E., ROC methodology in radiologic imaging. Invest Radiol21, 720–733 (1986).
12.
SwetsJ. A., Measuring the accuracy of diagnostic systems. Science240, 1285–1293 (1988).
13.
HanleyJ. A., McNeilB. J., The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology143, 29–36 (1982).
14.
JiangY., MetzC. E., NishikawaR. M., A receiver operating characteristic partial area index for highly sensitive diagnostic tests. Radiology201, 745–750 (1996).
15.
DorfmanD. D., BerbaumK. S., MetzC. E., Receiver operating characteristic rating analysis. Generalization to the population of readers and patients with the jackknife method. Invest Radiol27, 723–731 (1992).
16.
ElmoreJ. G., WellsC. K., LeeC. H., HowardD. H., FeinsteinA. R., Variability in radiologists' interpretations of mammograms. N Engl J Med331, 1493–1499 (1994).
17.
BeamC. A., LaydeP. M., SullivanD. C., Variability in the interpretation of screening mammograms by US radiologists. Findings from a national sample. Arch Intern Med156, 209–213 (1996).
18.
SchmidtR. A., NewsteadG. M., LinverM. N., EklundG. W., MetzC. E., WinklerM. N., NishikawaR. M., Mammographic screening sensitivity of general radiologists. In: Digital Mammography. pp. 383–388. Eds., KarssemeijerN., ThijssenM., HendriksJ., Van ErningL., Kluwer Academic Publishers, Dordrecht (1998).
19.
JiangY., NishikawaR. M., SchmidtR. A., ToledanoA. Y., DoiK., Potential of computer-aided diagnosis to reduce variability in radiologists' interpretations of mammograms depicting microcalcifications. Radiology220, 787–794 (2001).
20.
BeidenS. V., WagnerR. F., CampbellG., MetzC. E., JiangY., Components-of-variance models for random-effects ROC analysis: The case of unequal variance structures across modalities. Acad Radiol8, 605–615 (2001).
21.
CohenJ. A., coefficient of agreement for nominal scales. Educ Psychol Meas20, 37–46 (1960).
22.
HuoZ., GigerM. L., VybornyC. J., MetzC. E., Effectiveness of CAD in the diagnosis of breast cancer: An observer study on an independent database of mammograms. Radiology (in press).
23.
ChanH. P., DoiK., VybornyC. J., SchmidtR. A., MetzC. E., LamK. L., OguraT., WuY. Z., MacMahonH., Improvement in radiologists' detection of clustered microcalcifications on mammograms. The potential of computer-aided diagnosis. Invest Radiol25, 1102–1110 (1990).
24.
KegelmeyerW. P.Jr., PrunedaJ. M., BourlandP. D., HillisA., RiggsM. W., NipperM. L., Computer-aided mammographic screening for spiculated lesions. Radiology191, 331–337 (1994).
25.
FreerT. W., UlisseyM. J., Screening mammography with computer-aided detection: Prospective study of 12, 860 patients in a community breast center. Radiology220, 781–786 (2001).
26.
BoydN. F., LockwoodG. A., MartinL. J., KnightJ. A., JongR. A., FishellE., ByngJ. W., YaffeM. J., TritchlerD. L., Mammographic densities and risk of breast cancer among subjects with a family history of this disease. J Natl Cancer Inst91, 1404–1408 (1999).
27.
HuoZ., GigerM. L., WolvertonD. E., ZhongW., CummingS., OlopadeO. I., Computerized analysis of mammographic parenchymal patterns for breast cancer risk assessment: Feature selection. Med Phys27, 4–12 (2000).
