A knee osteoarthritis score prediction method and system fusing multi-modal image features

By integrating multimodal imaging features, this predictive method utilizes T1-weighted magnetic resonance imaging (MRI) sequences and clinical features to segment the popliteal artery region and extract radiomics and deep learning features. This addresses the insufficient accuracy of knee osteoarthritis prediction in existing technologies, achieving earlier and more accurate prediction results.

CN119741262BActive Publication Date: 2026-06-09SOUTHERN MEDICAL UNIVERSITY
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
SOUTHERN MEDICAL UNIVERSITY
Filing Date
2024-11-29
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing methods for predicting knee osteoarthritis rely on clinical symptoms and imaging examinations, and their accuracy and sensitivity need to be improved.

Method used

A prediction method that integrates multimodal imaging features acquires T1-weighted magnetic resonance imaging of the knee joint, segments the popliteal artery region, extracts radiomics and deep learning features, and combines them with clinical features to make predictions using a logistic regression classifier.

Benefits of technology

It improves the accuracy and reliability of predicting knee osteoarthritis, enabling earlier and more accurate prediction of the onset and progression of osteoarthritis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure GDA0005263543480000091
    Figure GDA0005263543480000091
  • Figure GDA0005263543480000161
    Figure GDA0005263543480000161
  • Figure GDA0005263543480000181
    Figure GDA0005263543480000181
Patent Text Reader

Abstract

The application discloses a knee osteoarthritis score prediction method and system fusing multi-modal image features, wherein the knee osteoarthritis score prediction method fusing multi-modal image features obtains a knee osteoarthritis prediction score of an object through five steps.The application has the beneficial effects that: 1. By fusing deep learning, image feature and clinical features, the knee osteoarthritis T1 sequence image is processed and analyzed, the popliteal artery ROI is marked, and the related image feature and deep network feature are extracted; compared with the traditional single feature analysis, the application provides more comprehensive and accurate disease evaluation, and improves the accuracy and reliability of the knee osteoarthritis prediction. 2. The application is based on the significant connection between the popliteal artery and the knee osteoarthritis, and innovatively analyzes the image features of the popliteal artery and combines the clinical data; the method can not only earlier and more accurately predict the occurrence and progress of osteoarthritis, but also significantly improve the prediction effect.
Need to check novelty before this filing date? Find Prior Art