An Explainable Contrastive-based Dilated Convolutional Network with Transformer for Pediatric Pneumonia Detection

Chandravardhan Singh Raghaw, Parth Shirish Bhore, Mohammad Zia Ur Rehman, Nagendra Kumar

Abstract: Highlights•Contrastive-inspired convolutional network with transformer for pneumonia detection.•Combined contrastive-entropy loss to improve feature learning and classifier learning.•Feature fusion strategy to capture joint global and fine-grained local information.•XCCNet enables deep insights for pneumonia classification through explainability.•Extensive evaluations of four datasets demonstrate the superiority of XCCNet.