Expert identification of visual primitives used by CNNs during mammogram classification

Jimmy Wu, Diondra Peck, Scott Hsieh, Vandana Dialani, Constance D. Lehman, Bolei Zhou, Vasilis Syrgkanis, Lester MacKey, Genevieve Patterson

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Scopus citations

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.

Original languageEnglish (US)
Title of host publicationMedical Imaging 2018
Subtitle of host publicationComputer-Aided Diagnosis
EditorsKensaku Mori, Nicholas Petrick
PublisherSPIE
ISBN (Electronic)9781510616394
DOIs
StatePublished - 2018
EventMedical Imaging 2018: Computer-Aided Diagnosis - Houston, United States
Duration: Feb 12 2018Feb 15 2018

Publication series

NameProgress in Biomedical Optics and Imaging - Proceedings of SPIE
Volume10575
ISSN (Print)1605-7422

Other

OtherMedical Imaging 2018: Computer-Aided Diagnosis
CountryUnited States
CityHouston
Period2/12/182/15/18

Keywords

  • deep learning for diagnosis
  • expert-in-the-loop methods
  • interpretable machine learning
  • Medical image understanding

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Biomaterials
  • Atomic and Molecular Physics, and Optics
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Expert identification of visual primitives used by CNNs during mammogram classification'. Together they form a unique fingerprint.

Cite this