POTHER: Patch-Voted Deep Learning-Based Chest X-ray Bias Analysis for COVID-19 Detection

TL;DR

This paper introduces POTHER, a deep learning method with explainability for COVID-19 detection in chest X-rays, addressing confounding factors to improve robustness and interpretability.

Contribution

The paper presents a novel patch-voting approach that reduces confounding influences in deep learning models for chest X-ray analysis, enhancing robustness and interpretability.

Findings

The proposed method outperforms previous models in robustness against confounders.

It achieves comparable accuracy to state-of-the-art methods.

The approach enhances explainability of model decisions.

Abstract

A critical step in the fight against COVID-19, which continues to have a catastrophic impact on peoples lives, is the effective screening of patients presented in the clinics with severe COVID-19 symptoms. Chest radiography is one of the promising screening approaches. Many studies reported detecting COVID-19 in chest X-rays accurately using deep learning. A serious limitation of many published approaches is insufficient attention paid to explaining decisions made by deep learning models. Using explainable artificial intelligence methods, we demonstrate that model decisions may rely on confounding factors rather than medical pathology. After an analysis of potential confounding factors found on chest X-ray images, we propose a novel method to minimise their negative impact. We show that our proposed method is more robust than previous attempts to counter confounding factors such as ECG leads in chest X-rays that often influence model classification decisions. In addition to being robust, our method achieves results comparable to the state-of-the-art. The source code and pre-trained weights are publicly available at (https://github.com/tomek1911/POTHER).