XAI for Skin Cancer Detection with Prototypes and Non-Expert Supervision

TL;DR

This paper presents an interpretable prototype-based model for skin cancer detection that incorporates non-expert supervision, leading to improved performance and interpretability without requiring expert input.

Contribution

It introduces a novel prototypical-part model guided by non-expert feedback, enhancing interpretability and accuracy in melanoma diagnosis.

Findings

Achieves superior performance over non-interpretable models

Ensures prototypes focus on relevant skin lesion areas

Operates effectively without expert supervision

Abstract

Skin cancer detection through dermoscopy image analysis is a critical task. However, existing models used for this purpose often lack interpretability and reliability, raising the concern of physicians due to their black-box nature. In this paper, we propose a novel approach for the diagnosis of melanoma using an interpretable prototypical-part model. We introduce a guided supervision based on non-expert feedback through the incorporation of: 1) binary masks, obtained automatically using a segmentation network; and 2) user-refined prototypes. These two distinct information pathways aim to ensure that the learned prototypes correspond to relevant areas within the skin lesion, excluding confounding factors beyond its boundaries. Experimental results demonstrate that, even without expert supervision, our approach achieves superior performance and generalization compared to non-interpretable models.