Scholarlink

Pneumonia is a crucial worldwide health issue and it is important to make the right diagnosis as soon as possible to start the right treatment. In this article, a predictive model based on Python in which CNNs are employed to determine and classify pneumonia from chest X-ray images is proposed. Using TensorFlow and Keras to design complex deep learning models, the proposed model performs the automation of pneumonia detection with fairly high accuracy. The CNN model was evaluated on a large dataset, and it provided an accuracy of 94.54%; thus, it can be used to distinguish between pneumonia-positive and normal images. The robustness of the model is also supported by other metrics such as recall, precision, and AUC-ROC that suggest that the model is capable of competing with current diagnosis practices and experienced radiologists. This article is significant in highlighting the revolutionary application of CNN in improving the current healthcare diagnosis especially in the detection of pneumonia and in the development of efficient and timely diagnosis systems. With the advancement in medical imaging and deep learning, this research article gives a significant contribution to enhancing patient care and the effectiveness of the healthcare delivery system.

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

To Improve Pneumonia Detection Accuracy by Combining the Strengths of Convolutional Neural Networks (CNN) and Random Forest Classifiers

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ajay Pal Singh,
  • Amarjeet Kumar,
  • Amit Kumar Jaiswal,
  • Ujjwal Rai,
  • Manav Agrawal

摘要

Pneumonia is a crucial worldwide health issue and it is important to make the right diagnosis as soon as possible to start the right treatment. In this article, a predictive model based on Python in which CNNs are employed to determine and classify pneumonia from chest X-ray images is proposed. Using TensorFlow and Keras to design complex deep learning models, the proposed model performs the automation of pneumonia detection with fairly high accuracy. The CNN model was evaluated on a large dataset, and it provided an accuracy of 94.54%; thus, it can be used to distinguish between pneumonia-positive and normal images. The robustness of the model is also supported by other metrics such as recall, precision, and AUC-ROC that suggest that the model is capable of competing with current diagnosis practices and experienced radiologists. This article is significant in highlighting the revolutionary application of CNN in improving the current healthcare diagnosis especially in the detection of pneumonia and in the development of efficient and timely diagnosis systems. With the advancement in medical imaging and deep learning, this research article gives a significant contribution to enhancing patient care and the effectiveness of the healthcare delivery system.