Feature Analysis of Knee Joint Magnetic Resonance Imaging Based on Artificial Intelligence

NCT07784686 | Completed

• 1 other identifier: K-HG-202607-01
• Study Type: observational
• Target: 9,500
• Locations: 1 country
• Sites: 1

Brief Summary

This retrospective observational study aims to investigate imaging features of the knee joint on magnetic resonance imaging (MRI) using artificial intelligence (AI)-based image analysis. Existing knee MRI examinations from eligible participants are retrospectively reviewed and analyzed. AI methods are used to identify and characterize anatomical structures and imaging abnormalities of the knee and to quantitatively evaluate relevant imaging features. The study aims to assess the feasibility and performance of AI-assisted MRI analysis and to explore its potential value in improving the objective and reproducible evaluation of knee joint imaging.

Conditions

Knee Joint

Keywords

Knee Joint

Outcome Measures

Primary Outcomes (1)

• Dice Similarity Coefficient for AI-Based Knee MRI Segmentation

  The Dice similarity coefficient will be used to evaluate the spatial agreement between artificial intelligence-generated segmentations and reference annotations. The Dice coefficient ranges from 0 to 1, with higher values indicating greater agreement.

  At completion of retrospective MRI image analysis

Study Arms (1)

Retrospective Knee MRI Cohort

Participants who underwent knee MRI examinations as part of routine clinical care and met the predefined eligibility criteria were retrospectively included. Existing MRI images and relevant clinical information were used for artificial intelligence-based image analysis. No intervention was assigned to participants for the purpose of this study.

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population consisted of patients who underwent knee MRI examinations at participating medical centers during the predefined study period. Eligible participants were retrospectively identified from existing imaging and clinical databases according to the predefined inclusion and exclusion criteria.

You may qualify if:

• Complete knee MRI data, including sagittal and coronal images. Adequate image quality without significant motion artifacts, metal artifacts, or magnetic susceptibility artifacts, with clear visualization of key anatomical structures of the knee.
• Complete demographic and clinical information available for study grouping and analysis.

You may not qualify if:

• Incomplete MRI data or poor image quality. History of knee surgery or implantation. Incomplete clinical information.

Sponsors & Collaborators

• Tang-Du Hospital: lead

Study Sites (1)

Tangdu Hospital, Fourth Military Medical University

Xi'an, Shaanxi, 710038, China

Location

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: August 21, 2026
• First Posted: August 25, 2026
• Study Start: July 2, 2026
• Primary Completion: July 20, 2026
• Study Completion: August 1, 2026
• Last Updated: August 25, 2026

Record last verified: 2026-08

Data Sharing

• IPD Sharing: Will not share

Locations

CN (1)