Scholarlink

In recent years, the integration of artificial intelligence in the medical field has changed the way healthcare professionals diagnose disease. Researchers have been using artificial intelligence technology to enhance the diagnostic process, especially through the analysis of medical imaging and clinical data. This study specifically investigated the effectiveness of two convolutional neural network models, VGG16 and VGG19, in predicting skin-related diseases. Through comprehensive evaluation, the study found that the VGG19 model performed significantly better than previous models, reaching an accuracy of 98.87% on the training data set and 98.22% on the test data set. These results demonstrate that VGG19 not only performs well in training but also maintains high performance when applied to unseen material. This suggests that VGG19 may become a valuable tool for clinicians, potentially improving diagnostic accuracy and patient outcomes. Ultimately, the findings using the VGG19 artificial intelligence model could hopefully enhance healthcare providers’ ability to diagnose skin-related diseases.

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

Enhancing Skin Disease Diagnosis Through Advanced Imaging Classification Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Ming-Ju Chen,
  • Pei-Yun Wang,
  • Han-Jie Shih,
  • Wen-Yen Chang,
  • Ci-An Lai,
  • Jia-Lang Xu

摘要

In recent years, the integration of artificial intelligence in the medical field has changed the way healthcare professionals diagnose disease. Researchers have been using artificial intelligence technology to enhance the diagnostic process, especially through the analysis of medical imaging and clinical data. This study specifically investigated the effectiveness of two convolutional neural network models, VGG16 and VGG19, in predicting skin-related diseases. Through comprehensive evaluation, the study found that the VGG19 model performed significantly better than previous models, reaching an accuracy of 98.87% on the training data set and 98.22% on the test data set. These results demonstrate that VGG19 not only performs well in training but also maintains high performance when applied to unseen material. This suggests that VGG19 may become a valuable tool for clinicians, potentially improving diagnostic accuracy and patient outcomes. Ultimately, the findings using the VGG19 artificial intelligence model could hopefully enhance healthcare providers’ ability to diagnose skin-related diseases.