This study introduces a computer-aided system for detecting ulcerative colitis, a chronic inflammatory bowel disease. The system can tell the various ulcer colitis severity levels by using a pre-trained deep learning model called EfficienetB2V3 and fine-tuning it on a special medical image dataset called LIMUC. This approach, known as transfer learning, improves the model’s accuracy by leveraging knowledge from a vast amount of data. The resulting system achieved 91% training accuracy in classifying ulcer colitis images and is designed to support healthcare providers in making earlier and more accurate ulcer colitis diagnoses.

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

Ulcerative Colitis Diagnosis Through Efficient Net Transfer Learning

  • Dharmendra Gupta,
  • Jayesh Gangrade,
  • Yadvendra Pratap Singh

摘要

This study introduces a computer-aided system for detecting ulcerative colitis, a chronic inflammatory bowel disease. The system can tell the various ulcer colitis severity levels by using a pre-trained deep learning model called EfficienetB2V3 and fine-tuning it on a special medical image dataset called LIMUC. This approach, known as transfer learning, improves the model’s accuracy by leveraging knowledge from a vast amount of data. The resulting system achieved 91% training accuracy in classifying ulcer colitis images and is designed to support healthcare providers in making earlier and more accurate ulcer colitis diagnoses.