Interpretability-Guided Inductive Bias For Deep Learning Based Medical Image

Deep learning methods provide state of the art performance for supervised learning based medical image analysis. However it is essential that trained models extract clinically relevant features for downstream tasks as, otherwise, shortcut learning and generalization issues can occur. Furthermore in the medical field, trustability and transparency of current deep learning systems is a much desired property. In this paper we propose an interpretability-guided inductive bias approach enforcing that learned features yield more distinctive and spatially consistent saliency maps for different class labels of trained models, leading to improved model performance. We achieve our objectives by incorporating a class-distinctiveness loss and a spatial-consistency regularization loss term. Experimental results for medical image classification and segmentation tasks show our proposed approach outperforms conventional methods, while yielding saliency maps in higher agreement with clinical experts. Additionally, we show how information from unlabeled images can be used to further boost performance. In summary, the proposed approach is modular, applicable to existing network architectures used for medical imaging applications, and yields improved learning rates, model robustness, and model interpretability.

Automated summary

In this paper, we propose an interpretabilit-guided inductive bias approach that enhances deep learning models in medical image analysis by enforcing the extraction of clinically relevant and spatially consistent features. Through experiments on medical image classification and segmentation tasks, we demonstrate that our approach outperforms conventional methods and generates saliency maps in higher agreement with clinical experts.