Interpretable breast cancer classification using CNNs on mammographic images
This work tackles interpretability in breast cancer classification for clinical practice, but it is incremental as it compares existing methods without introducing new ones.
The study compared post-hoc interpretability techniques LIME, Grad-CAM, and Kernel SHAP for CNNs on mammographic images to address black-box concerns, finding that Grad-CAM provided comprehensive insights into patterns for normal, benign, and malignant tissue.
Deep learning models have achieved promising results in breast cancer classification, yet their 'black-box' nature raises interpretability concerns. This research addresses the crucial need to gain insights into the decision-making process of convolutional neural networks (CNNs) for mammogram classification, specifically focusing on the underlying reasons for the CNN's predictions of breast cancer. For CNNs trained on the Mammographic Image Analysis Society (MIAS) dataset, we compared the post-hoc interpretability techniques LIME, Grad-CAM, and Kernel SHAP in terms of explanatory depth and computational efficiency. The results of this analysis indicate that Grad-CAM, in particular, provides comprehensive insights into the behavior of the CNN, revealing distinctive patterns in normal, benign, and malignant breast tissue. We discuss the implications of the current findings for the use of machine learning models and interpretation techniques in clinical practice.