Scholarlink

Osteoarthritis (OA) is the most widespread kind of arthritis and is especially common in the knee, severely crippling people worldwide. When it comes to correctly diagnosing and treating musculoskeletal diseases, medical imaging is essential. Osteoarthritis in the knee can cause years of disability. Unfortunately, early sickness progression identification is hampered by the very subjective nature of the standard procedures. Nevertheless, the manual procedure takes along time and is tedious. Performance is not always optimized. This study employed a deep learning approach to detect issues early on. This approach uses the CNN model to determine if the patient has osteoarthritis in their knees or is in good health. The OAI dataset, which comes from Kaggle, is utilized by this method. The result analysis of method accuracy is 92.9%, F1 score: 95.4 Recall is 95.04 and Precision: 95.8. This yields the best outcome from the earlier techniques. This research primarily focuses on the binary classification of an individual’s status as either diseased or not. We will continue to work on KOA severity detection in the future.

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

Automated KOA Detection in X-ray Images Using Convolutional Neural Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Suman Rani,
  • Minakshi Memoria,
  • Kapil Joshi

摘要

Osteoarthritis (OA) is the most widespread kind of arthritis and is especially common in the knee, severely crippling people worldwide. When it comes to correctly diagnosing and treating musculoskeletal diseases, medical imaging is essential. Osteoarthritis in the knee can cause years of disability. Unfortunately, early sickness progression identification is hampered by the very subjective nature of the standard procedures. Nevertheless, the manual procedure takes along time and is tedious. Performance is not always optimized. This study employed a deep learning approach to detect issues early on. This approach uses the CNN model to determine if the patient has osteoarthritis in their knees or is in good health. The OAI dataset, which comes from Kaggle, is utilized by this method. The result analysis of method accuracy is 92.9%, F1 score: 95.4 Recall is 95.04 and Precision: 95.8. This yields the best outcome from the earlier techniques. This research primarily focuses on the binary classification of an individual’s status as either diseased or not. We will continue to work on KOA severity detection in the future.