Breast Cancer Diagnosis and Prognosis via Linear Programming

O. L. Mangasarian, W. N. Street and W. H. Wolberg

This ongoing multi-disciplinary research directly addresses problems arising in the diagnosis and treatment of breast cancer. Early detection of breast cancer is enhanced and unnecessary surgery avoided by diagnosing breast masses from Fine Needle Aspirates (FNA’s). using and extending results from the fields of optimization, machine learning, statistics and image processing, a software system was created that allows highly accurate diagnosis of breast FNA’s even by untrained users. The system is in current use at the University of Wisconsin Hospitals. For malignant cases, treatment decisions are enhanced by accurately predicting the long term behavior of the disease. This paper summarizes our recent work in both areas of diagnosis and prognosis, with emphasis on the more difficult latter problem.


This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.