Explainable Image Classification with Reduced Overconfidence for Tissue Characterisation

TL;DR

This paper introduces a novel explainability method for image classification that incorporates risk estimation into pixel attribution, improving interpretability and confidence in tissue characterization models, validated on medical and standard datasets.

Contribution

It is the first to integrate risk estimation with pixel attribution for enhanced explainability in tissue image classification models.

Findings

Outperforms state-of-the-art explainability methods on pCLE and ImageNet datasets.

Provides pixel-wise risk estimation alongside attribution maps.

Improves confidence and interpretability of tissue classification models.

Abstract

The deployment of Machine Learning models intraoperatively for tissue characterisation can assist decision making and guide safe tumour resections. For image classification models, pixel attribution methods are popular to infer explainability. However, overconfidence in deep learning model's predictions translates to overconfidence in pixel attribution. In this paper, we propose the first approach which incorporates risk estimation into a pixel attribution method for improved image classification explainability. The proposed method iteratively applies a classification model with a pixel attribution method to create a volume of PA maps. This volume is used for the first time, to generate a pixel-wise distribution of PA values. We introduce a method to generate an enhanced PA map by estimating the expectation values of the pixel-wise distributions. In addition, the coefficient of variation (CV) is used to estimate pixel-wise risk of this enhanced PA map. Hence, the proposed method not only provides an improved PA map but also produces an estimation of risk on the output PA values. Performance evaluation on probe-based Confocal Laser Endomicroscopy (pCLE) data and ImageNet verifies that our improved explainability method outperforms the state-of-the-art.