The applicants proposed to investigate the feasibility of developing a Case Based Reasoning (CBR) system to provide support for the clinical decision to perform breast biopsy. The system is designed to aid in the decision to biopsy those patients that have suspicious mammographic findings. The decision to biopsy is a two stage process: 1) mammographer views the mammogram and determines the presence or absence of image features such as calcifications and masses, 2) these features are merged to form a diagnosis. This work is motivated by a recent study which found that 52% of missed breast cancers are due to errors at the decision step and by the fact that about 80% of the biopsies that are performed are benign. The applicants proposed a system to significantly improve this performance by a CBR approach that utilizes a large database of cases with known outcomes. The innovative research would be the development and implementation of statistical measures for deciding which cases in the database come enough to matching the test case. The clinician reads a mammogram and enters the findings into a computer using a standard reporting lexicon (BI-RADS). The database is searched for similar cases and the fraction that were malignant is returned. This likelihood of malignancy is an intuitive response which the clinician can then include in the medical decision.
