WEEP: A method for spatial interpretation of weakly supervised CNN models in computational pathology
This addresses the need for improved interpretability in computational pathology for research and diagnostic applications, but it is an incremental method focused on a specific domain.
The authors tackled the problem of spatial interpretability in weakly supervised CNN models for computational pathology by proposing WEEP, a method that identifies the spatial area of whole-slide images needed for predictions, demonstrating it on a breast cancer classification task.
Deep learning enables the modelling of high-resolution histopathology whole-slide images (WSI). Weakly supervised learning of tile-level data is typically applied for tasks where labels only exist on the patient or WSI level (e.g. patient outcomes or histological grading). In this context, there is a need for improved spatial interpretability of predictions from such models. We propose a novel method, Wsi rEgion sElection aPproach (WEEP), for model interpretation. It provides a principled yet straightforward way to establish the spatial area of WSI required for assigning a particular prediction label. We demonstrate WEEP on a binary classification task in the area of breast cancer computational pathology. WEEP is easy to implement, is directly connected to the model-based decision process, and offers information relevant to both research and diagnostic applications.