Scholarlink

The ability of deep learning models to identify intricate patterns and characteristics from MRIs, CT scans, and X-rays is crucial for the timely detection of tuberculosis, pneumonia, and lung diseases. For the diagnosis and treatment of diseases, X-ray images are essential. This work assesses multiple deep learning algorithms and image processing approaches to determine the best approach for X-ray image-based pneumonia prediction. It also incorporates image enhancement techniques to improve the prediction. Among the deep learning models that were assessed are MobileNet, ResNet, EfficientNet, VGG16, and VGG19. These models’ accuracy in predicting diseases varied: EfficientNetB3 had the highest accuracy at 97.6%, while MobileNetV2 scored 88.9%, VGG16 scored 95%, VGG19 scored 86%, and ResNet50 scored 91%. EfficientNetB3 outperforms the other models, according to the results. The paper examines the potential applications of image processing in healthcare and highlights the importance of utilizing deep learning systems to enhance disease detection.

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

A Deep Learning Approach to Pneumonia Prediction Through X-Ray Image Analysis

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jennie Gratia Franklin,
  • K. N. Sengamali,
  • P. Divya,
  • P. Prakash,
  • P. Kasthuri

摘要

The ability of deep learning models to identify intricate patterns and characteristics from MRIs, CT scans, and X-rays is crucial for the timely detection of tuberculosis, pneumonia, and lung diseases. For the diagnosis and treatment of diseases, X-ray images are essential. This work assesses multiple deep learning algorithms and image processing approaches to determine the best approach for X-ray image-based pneumonia prediction. It also incorporates image enhancement techniques to improve the prediction. Among the deep learning models that were assessed are MobileNet, ResNet, EfficientNet, VGG16, and VGG19. These models’ accuracy in predicting diseases varied: EfficientNetB3 had the highest accuracy at 97.6%, while MobileNetV2 scored 88.9%, VGG16 scored 95%, VGG19 scored 86%, and ResNet50 scored 91%. EfficientNetB3 outperforms the other models, according to the results. The paper examines the potential applications of image processing in healthcare and highlights the importance of utilizing deep learning systems to enhance disease detection.