XAI-Guided Analysis of Residual Networks for Interpretable Pneumonia Detection in Paediatric Chest X-rays
This addresses the need for fast and accurate diagnostic tools for pneumonia in children, which is a leading cause of death, though it is incremental in applying existing methods to a specific medical domain.
The paper tackled the problem of diagnosing paediatric pneumonia in chest X-rays by proposing an interpretable deep learning model based on ResNets, achieving high classification accuracy (95.94%), AUC-ROC (98.91%), and Cohen's Kappa (0.913) with clinically meaningful visual explanations.
Pneumonia remains one of the leading causes of death among children worldwide, underscoring a critical need for fast and accurate diagnostic tools. In this paper, we propose an interpretable deep learning model on Residual Networks (ResNets) for automatically diagnosing paediatric pneumonia on chest X-rays. We enhance interpretability through Bayesian Gradient-weighted Class Activation Mapping (BayesGrad-CAM), which quantifies uncertainty in visual explanations, and which offers spatial locations accountable for the decision-making process of the model. Our ResNet-50 model, trained on a large paediatric chest X-rays dataset, achieves high classification accuracy (95.94%), AUC-ROC (98.91%), and Cohen's Kappa (0.913), accompanied by clinically meaningful visual explanations. Our findings demonstrate that high performance and interpretability are not only achievable but critical for clinical AI deployment.