Related Experiment Video
Updated: Jun 3, 2025
07:32
Standardized Histomorphometric Evaluation of Osteoarthritis in a Surgical Mouse Model
Published on: May 6, 2020
12.0K
Orthopedic surgeon level joint angle assessment with artificial intelligence based on photography: a pilot study
Seung Min Ryu1, Keewon Shin2, Chang Hyun Doh3
1Department of Orthopedic Surgery, Seoul Medical Center, Seoul, 02053, South Korea.
Biomedical Engineering Letters
|January 9, 2025
Summary
This study introduces an AI-based method using clinical photos to measure shoulder range of motion (ROM), offering improved accuracy and consistency over traditional goniometry for internal rotation assessment.
Area of Science:
- Biomedical Engineering
- Artificial Intelligence in Medicine
- Musculoskeletal Imaging
Background:
- Accurate shoulder range of motion (ROM) assessment is vital for patient progress evaluation.
- Traditional manual goniometry presents limitations in precision and inter-observer variability, particularly for shoulder internal rotation (IR).
Purpose of the Study:
- To introduce and validate an artificial intelligence (AI)-based approach for quantifying shoulder and elbow ROM using clinical photography.
- To assess the accuracy and reliability of the AI algorithm compared to human observers.
Main Methods:
- An AI model (MMPose with HR-NET) analyzed 150 clinical photographs to detect 17 anatomical landmarks.
- A random forest classifier (PoseRF) categorized poses, and ROM angles were calculated.
- AI-derived measurements were correlated with concurrent manual measurements by two clinicians.
Main Results:
- The AI algorithm achieved high accuracy in landmark detection (96% shoulder, 100% elbow) and pose detection (95% overall).
- Intraclass correlation coefficients (ICCs) between AI and human observers ranged from 0.965 to 0.997, indicating excellent reliability.
- No statistically significant differences were found between AI and human measurements (p > 0.05).
Conclusions:
- The AI-based algorithm demonstrates performance comparable to human observers in quantifying shoulder and elbow ROM from clinical photographs.
- This AI approach shows potential for enhanced consistency in shoulder internal rotation measurement compared to traditional methods.
