Scholarlink

The lungs are the main organ of the respiratory system, which also consists of the nostrils, the mouth, the throat, the larynx bronchial tubes and pulmonary arteries. The major causes of death are due to the respiratory diseases, which include ailments including pneumonia, asthma, and chronic bronchitis. The chest x-rays (CXR) to estimate dynamic parameters like pulmonary function is common, affordable, and fundamental screening technique utilized for static examinations of organic disorders and physical anomalies. This study is to assess the performance of the Deep Learning (DL).architectures like CovCXR-Net, Deep CCXR, CNN-O-ELMNet for classifying and predicting lung diseases from X-rays. The methods are implemented in python and evaluated with CXR dataset, a publically available dataset. The result findings show that the CNN-O-ELMNet outperforms with the accuracy of 97.8%. This assessment study will be very helpful for clinicians, enhancing the diagnosis and treatment of lung disorders from X-rays.

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

Performance Evaluation of Deep Learning Models for the Classification of Lung Diseases in X-Ray Images

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • A. Balaji,
  • S. Brintha Rajakumari

摘要

The lungs are the main organ of the respiratory system, which also consists of the nostrils, the mouth, the throat, the larynx bronchial tubes and pulmonary arteries. The major causes of death are due to the respiratory diseases, which include ailments including pneumonia, asthma, and chronic bronchitis. The chest x-rays (CXR) to estimate dynamic parameters like pulmonary function is common, affordable, and fundamental screening technique utilized for static examinations of organic disorders and physical anomalies. This study is to assess the performance of the Deep Learning (DL).architectures like CovCXR-Net, Deep CCXR, CNN-O-ELMNet for classifying and predicting lung diseases from X-rays. The methods are implemented in python and evaluated with CXR dataset, a publically available dataset. The result findings show that the CNN-O-ELMNet outperforms with the accuracy of 97.8%. This assessment study will be very helpful for clinicians, enhancing the diagnosis and treatment of lung disorders from X-rays.