LungX-Net: A CNN-Based Hybrid System for Multi-Disease Diagnosis in Chest X-Rays

View at publisher Summarise with VIRAT
Authors: Yuvasri S, Keerthi Dayalan, Nikita Soren
Cited by: 0
References: 14

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The proposed framework is primarily designed to classify three diseases, namely pneumonia, COVID-19, and tuberculosis, based on chest X-ray images. The dataset was downloaded from Kaggle. The proposed model is based on EfficientNet-B1, a CNN architecture that is scaled in depth, width, and resolution. This work focuses on developing a practical and reliable system that can support timely clinical decision-making through accurate multi-class lung disease detection. EfficientNet-B1 is better than traditional CNNs such as Visual Geometry Group 16 (VGG16) and Residual Neural Network (ResNet). It can achieve higher accuracy with far fewer parameters and requires a lower computational cost. This makes it suitable for medical image classification. The proposed system achieved an overall accuracy of 98% and class-wise F1-scores of 1.00, 0.83, 0.92, and 0.99 for COVID-19, Normal, Pneumonia, and Tuberculosis, respectively. Explainable AI (XAI) techniques are used to explain the reason for predictions.

Fields of Study & Subjects

COVID-19 diagnosis using AI Radiomics and Machine Learning in Medical Imaging Artificial Intelligence in Healthcare and Education

Publication details

Journal / source
International Conference on Computing, Communication, Control and Cyber-Physical Systems (I5CPS)
Pages
1–6
Publication date
26 February 2026
Language
EN