- Home
- Current Issue
- Article
Detecting Pneumonia from Chest X-Rays Using Convolutional Neural Network Model
Authors
1 2 3 4 5 RV College of Engineering, Bengaluru, Karnataka, India.
Article Information
- DOI
- No DOI
- Published In
- Volume 5, Issue 3 (2026)
- Published Online
- September-December 2026
- Pages
- 382-388
- Article Type
- Original Article
- Check for updates
- No DOI available for cross-reference
Abstract
View PDFPneumonia is a leading cause of mortality all across the globe. The lack of skilled medical expertise in regions where radiology expertise is limited exacerbates the problem. The most common way to detect pneumonia is via atrained medical observation and analysis of chest X-rays. Inconsistencies in interpretation and shortage of skilled radiologists pose a significant challenge. The aim of this study is to present a novel deep learning-based approach for auto detection of pneumonia with a high level of accuracy. In this approach we utilize the DenseNet121 architecture with transfer learning. The proposed methodology employs a systematic multi-stage process encompassing automated data loading, real-time augmentation, class imbalance handling, and network optimization. A publicly available chest X-ray dataset was processed using TensorFlow and Keras. The trained model achieved high predictive performance of 91.0%, an AUC score of 0.9511, a F1 score of 0.9279 at optimal threshold of 0.68, demonstrating its capability to distinguish pneumonia from normal cases with strong generalization on unseen data. Future directions include interface development for real-time diagnostics and expansion to multi-class classification of thoracic diseases.
Related Articles
Share Article
*Instagram doesn't support direct link sharing from web. Copy the link and share it in your Instagram story or post.