English

DeepMiner: Discovering Interpretable Representations for Mammogram Classification and Explanation

Computer Vision and Pattern Recognition 2021-11-02 v2

Abstract

We propose DeepMiner, a framework to discover interpretable representations in deep neural networks and to build explanations for medical predictions. By probing convolutional neural networks (CNNs) trained to classify cancer in mammograms, we show that many individual units in the final convolutional layer of a CNN respond strongly to diseased tissue concepts specified by the BI-RADS lexicon. After expert annotation of the interpretable units, our proposed method is able to generate explanations for CNN mammogram classification that are consistent with ground truth radiology reports on the Digital Database for Screening Mammography. We show that DeepMiner not only enables better understanding of the nuances of CNN classification decisions but also possibly discovers new visual knowledge relevant to medical diagnosis.

Keywords

Cite

@article{arxiv.1805.12323,
  title  = {DeepMiner: Discovering Interpretable Representations for Mammogram Classification and Explanation},
  author = {Jimmy Wu and Bolei Zhou and Diondra Peck and Scott Hsieh and Vandana Dialani and Lester Mackey and Genevieve Patterson},
  journal= {arXiv preprint arXiv:1805.12323},
  year   = {2021}
}

Comments

Harvard Data Science Review (HDSR), 2021. Code available at https://github.com/jimmyyhwu/ddsm-visual-primitives

R2 v1 2026-06-23T02:14:18.480Z