Combining Similarity and Adversarial Learning to Generate Visual Explanation: Application to Medical Image Classification
This work addresses the critical need for explainable AI in medical imaging to build clinician confidence, offering an incremental improvement over existing perturbation-based explanation methods.
This paper introduces a method for generating visual explanations for black-box medical image classifiers by training two generators to create 'similar' and 'adversarial' images. The explanation is derived from the difference between these generated images, and the method reportedly outperforms state-of-the-art approaches on a large chest X-ray database.
Explaining decisions of black-box classifiers is paramount in sensitive domains such as medical imaging since clinicians confidence is necessary for adoption. Various explanation approaches have been proposed, among which perturbation based approaches are very promising. Within this class of methods, we leverage a learning framework to produce our visual explanations method. From a given classifier, we train two generators to produce from an input image the so called similar and adversarial images. The similar image shall be classified as the input image whereas the adversarial shall not. Visual explanation is built as the difference between these two generated images. Using metrics from the literature, our method outperforms state-of-the-art approaches. The proposed approach is model-agnostic and has a low computation burden at prediction time. Thus, it is adapted for real-time systems. Finally, we show that random geometric augmentations applied to the original image play a regularization role that improves several previously proposed explanation methods. We validate our approach on a large chest X-ray database.