Explaining Digital Pathology Models via Clustering Activations
This work addresses the need for better explainability in digital pathology to increase clinical adoption, though it appears incremental as it builds on existing clustering and visualization approaches.
The paper tackles the problem of explaining digital pathology models by introducing a clustering-based technique that reveals global model behavior and fine-grained information, unlike saliency methods focused on single slides, and demonstrates its usefulness on a prostate cancer detection model.
We present a clustering-based explainability technique for digital pathology models based on convolutional neural networks. Unlike commonly used methods based on saliency maps, such as occlusion, GradCAM, or relevance propagation, which highlight regions that contribute the most to the prediction for a single slide, our method shows the global behaviour of the model under consideration, while also providing more fine-grained information. The result clusters can be visualised not only to understand the model, but also to increase confidence in its operation, leading to faster adoption in clinical practice. We also evaluate the performance of our technique on an existing model for detecting prostate cancer, demonstrating its usefulness.