Share Email Print
cover

Proceedings Paper

Expert identification of visual primitives used by CNNs during mammogram classification
Author(s): Jimmy Wu; Diondra Peck; Scott Hsieh; Vandana Dialani; Constance D. Lehman; Bolei Zhou; Vasilis Syrgkanis; Lester Mackey; Genevieve Patterson
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

This work interprets the internal representations of deep neural networks trained for classification of diseased tissue in 2D mammograms. We propose an expert-in-the-loop inter- pretation method to label the behavior of internal units in convolutional neural networks (CNNs). Expert radiologists identify that the visual patterns detected by the units are correlated with meaningful medical phenomena such as mass tissue and calcificated vessels. We demonstrate that several trained CNN models are able to produce explanatory descriptions to support the final classification decisions. We view this as an important first step toward interpreting the internal representations of medical classification CNNs and explaining their predictions.

Paper Details

Date Published: 27 February 2018
PDF: 9 pages
Proc. SPIE 10575, Medical Imaging 2018: Computer-Aided Diagnosis, 105752T (27 February 2018); doi: 10.1117/12.2293890
Show Author Affiliations
Jimmy Wu, Massachusetts Institute of Technology (United States)
Diondra Peck, Harvard Univ. (United States)
Scott Hsieh, Univ. of California, Los Angeles (United States)
Vandana Dialani, Beth Israel Deaconess Medical Ctr. (United States)
Constance D. Lehman, Massachusetts General Hospital (United States)
Bolei Zhou, Massachusetts Institute of Technology (United States)
Vasilis Syrgkanis, Microsoft Research New England (United States)
Lester Mackey, Microsoft Research New England (United States)
Genevieve Patterson, Microsoft Research New England (United States)


Published in SPIE Proceedings Vol. 10575:
Medical Imaging 2018: Computer-Aided Diagnosis
Nicholas Petrick; Kensaku Mori, Editor(s)

© SPIE. Terms of Use
Back to Top