Enhanced Chest Disease Classification Using an Improved CheXNet Framework with EfficientNetV2-M and Optimization-Driven Learning

2025-12-08

AI Summary

To address diagnostic delays in chest radiography caused by radiologist shortages in primary healthcare settings, this paper proposes a lightweight multi-disease classification method tailored for resource-constrained environments, targeting five critical categories: cardiomegaly, COVID-19, normal, pneumonia, and tuberculosis. The method integrates an EfficientNetV2-M backbone with automatic mixed precision (AMP), AdamW optimization, cosine annealing learning rate scheduling, and exponential moving average (EMA) regularization to establish a highly efficient and stable training paradigm. Despite a 6.8× increase in model parameters, training time is reduced by 11.4% and training stability improves by 22.7%. On the test set, the model achieves an overall accuracy of 96.45% (p < 0.001) and a macro-F1 score of 91.08%, with exceptional detection performance for COVID-19 (99.95% accuracy) and tuberculosis (99.97% accuracy).

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Computer Vision: Multi-modal Vision
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

The interpretation of Chest X-ray is an important diagnostic issue in clinical practice and especially in the resource-limited setting where the shortage of radiologists plays a role in delayed diagnosis and poor patient outcomes. Although the original CheXNet architecture has shown potential in automated analysis of chest radiographs, DenseNet-121 backbone is computationally inefficient and poorly single-label classifier. To eliminate such shortcomings, we suggest a better classification framework of chest disease that relies on EfficientNetV2-M and incorporates superior training approaches such as Automatic Mixed Precision training, AdamW, Cosine Annealing learning rate scheduling, and Exponential Moving Average regularization. We prepared a dataset of 18,080 chest X-ray images of three source materials of high authority and representing five key clinically significant disease categories which included Cardiomegaly, COVID-19, Normal, Pneumonia, and Tuberculosis. To achieve statistical reliability and reproducibility, nine independent experimental runs were run. The suggested architecture showed significant gains with mean test accuracy of 96.45 percent compared to 95.30 percent at baseline (p less than 0.001) and macro-averaged F1-score increased to 91.08 percent (p less than 0.001). Critical infectious diseases showed near-perfect classification performance with COVID-19 detection having 99.95 percent accuracy and Tuberculosis detection having 99.97 percent accuracy. Although 6.8 times more parameters are included, the training time was reduced by 11.4 percent and performance stability was increased by 22.7 percent. This framework presents itself as a decision-support tool that can be used to respond to a pandemic, screen tuberculosis, and assess thoracic disease regularly in various healthcare facilities.

Problem

• Improving automated chest X-ray classification for clinical diagnosis
• Enhancing computational efficiency and accuracy in disease detection
• Addressing radiologist shortages with AI-driven diagnostic support

Innovation

• Replaced DenseNet-121 backbone with EfficientNetV2-M
• Used optimization techniques like AdamW and Cosine Annealing
• Applied Automatic Mixed Precision training to reduce time