100 Percent Accuracy in Breast Cancer Biopsy Analysis by Deep-Learning Computer Network
June 01, 2017
A deep-learning computer network developed through a collaboration of Case Western Reserve University, University of Pennsylvania, and University of Buffalo researchers has reached 100% accuracy in determining whether invasive forms of breast cancer were present in whole biopsy slides. The study, "Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent," was published in Scientific Reports. Penn researcher Dr. Michael Feldman, director the informatics division, and former Chair of Penn's Department of Pathology of Laboratory Medicine Dr. John Tomaszewski are co-authors. Dr. Tomaszewski, Chair in the Department of Pathology and Anatomical Sciences in the Jacobs School of Medicine and Biomedical Sciences at SUNY Buffalo, is internationally renowned for his research in pathology and prognostic factors in cancer, and development of quantitative image analysis tools used in digital pathology and automated cancer diagnostics. wHe as recently promoted to SUNY Distinguished Professor, the highest faculty rank in the SUNY system.