Scholarlink

<p>The COVID-19 pandemic has posed significant challenges for the timely detection of pneumonia and the effective management of virus-induced lung infections. Chest X-ray imaging has emerged as a crucial diagnostic tool due to its wide availability and low cost, yet manual interpretation is both time-consuming and reliant on expert precision. This study investigates the use of Convolutional Neural Networks (CNNs) for the automated classification of chest X-ray images to detect COVID-19 pneumonia. Several models, including traditional Neural Networks, Support Vector Machines (SVM), and various CNN architectures, were evaluated for their ability to distinguish between normal, viral pneumonia, and COVID-19 pneumonia cases. Among them, the VGG-19 network achieved the highest performance, with a test accuracy of 97.5% and a validation accuracy of 99.4%, while ResNet-50 also demonstrated strong results with a test accuracy of 94.16% and a validation accuracy of 95%. These findings highlight the potential of CNN-based systems to enhance early screening processes, reduce diagnostic delays, and support clinical decision-making. Future advancements can be realized by deploying more sophisticated CNN models on larger and more diverse datasets to improve diagnostic accuracy and reliability further.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning Models for Accurate Detection of COVID-19 Pneumonia from Chest X-Ray Images

  • 期刊论文

  • 发表日期：

    2025年8月21日

查看全文
  • C. P. Vijay,
  • R. Thejaswini,
  • Shalini Hanok,
  • J. Rajeshwari,
  • B. Madhu,
  • K. Prabhavathi,
  • G. R. Yathiraj

摘要

The COVID-19 pandemic has posed significant challenges for the timely detection of pneumonia and the effective management of virus-induced lung infections. Chest X-ray imaging has emerged as a crucial diagnostic tool due to its wide availability and low cost, yet manual interpretation is both time-consuming and reliant on expert precision. This study investigates the use of Convolutional Neural Networks (CNNs) for the automated classification of chest X-ray images to detect COVID-19 pneumonia. Several models, including traditional Neural Networks, Support Vector Machines (SVM), and various CNN architectures, were evaluated for their ability to distinguish between normal, viral pneumonia, and COVID-19 pneumonia cases. Among them, the VGG-19 network achieved the highest performance, with a test accuracy of 97.5% and a validation accuracy of 99.4%, while ResNet-50 also demonstrated strong results with a test accuracy of 94.16% and a validation accuracy of 95%. These findings highlight the potential of CNN-based systems to enhance early screening processes, reduce diagnostic delays, and support clinical decision-making. Future advancements can be realized by deploying more sophisticated CNN models on larger and more diverse datasets to improve diagnostic accuracy and reliability further.