Computerized detection and characterization of mass lesions in digital mammography
We are developing computer-aided diagnosis schemes for the detection of breast lesions in digital mammograms. The detection scheme employs a bilateral-subtraction technique that is based on the normal symmetry between the left and right breast images. The presence of asymmetries indicates a possible lesion. Feature-analysis techniques are then used to reduce the number of false-positive detections. In addition, characterization methods are being developed to aid in the benign/malignant decision. We plan to incorporate the schemes into a dedicated "intelligent" workstation for use as a "second opinion" in a mammographic screening program. Output from the computer would be displayed as an aid, leaving the final diagnostic decision with the radiologist.