Automated Detection of Pediatric Pneumonia from Chest X-Ray Images Using Deep Learning Models

Abstract

Pneumonia, a severe respiratory infection with high mortality rates among both infants and the elderly, presents a significant challenge in achieving early and accurate diagnosis. Human-assisted diagnosis is limited by factors such as expert availability, associated costs, and subjectivity in radiologist interpretations. To address these challenges, there is a pressing need for automated methods to detect pneumonia from X-rays. In this study, we propose and evaluate deep learning models, comprising custom-built CNN and transfer learning models (VGG16, Inception V3, and ResNet 152 V2), for the automated detection of pediatric pneumonia from radiographic images. Our experimental findings highlight the effectiveness of the VGG16 model, achieving an impressive accuracy of 92.63% and a high recall score of 97.18%, demonstrating its potential to enhance pneumonia detection.