Scholarlink

Chronic diseases are on rise in recent times and symptoms of these disorders need to be detected at an early stage. Traditional methods in handling these risks are quite resource intensive and costly with less accuracy. This paper explores transfer learning to address the challenge of limited data in diagnosing chronic diseases like chronic kidney disease, breast cancer, hepatitis, and Alzheimer’s. We propose a multi-modal framework utilizing pre-trained models: image recognition for medical images and natural language processing for textual data. Transfer learning aims to improve diagnostic accuracy and reduce training time, enabling development of adaptable tools for various chronic diseases. The implementation results show the performance of the model is promising generating an accuracy rate of 94%. Also the model gave a mean accuracy of 93.3% when tested with different chronic disorders.

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

Chronic Disease Diagnosis Using Multi-modal Transfer Learning Model

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Debanksh Guha,
  • Divya Avtaran,
  • Vandana Sharma,
  • Sushruta Mishra,
  • Rahul Lenka,
  • Ahmed Alkhayyat

摘要

Chronic diseases are on rise in recent times and symptoms of these disorders need to be detected at an early stage. Traditional methods in handling these risks are quite resource intensive and costly with less accuracy. This paper explores transfer learning to address the challenge of limited data in diagnosing chronic diseases like chronic kidney disease, breast cancer, hepatitis, and Alzheimer’s. We propose a multi-modal framework utilizing pre-trained models: image recognition for medical images and natural language processing for textual data. Transfer learning aims to improve diagnostic accuracy and reduce training time, enabling development of adaptable tools for various chronic diseases. The implementation results show the performance of the model is promising generating an accuracy rate of 94%. Also the model gave a mean accuracy of 93.3% when tested with different chronic disorders.