Related Experiment Video
Updated: Oct 10, 2025
08:19
Author Spotlight: Unraveling the Mechanobiology of Tendon Impingement – A Multiaxial Murine Hind Limb Explant Model
Published on: December 8, 2023
1.1K
Machine Learning-Based Ultrasomics for Predicting Subacromial Impingement Syndrome Stages
Hao Jiang1,2, Ling Chen3, Yu-Jia Zhao1
1Department of Medical Ultrasound, The First Affiliated Hospital of Guangxi Medical University, Nanning, China.
Summary
Machine learning ultrasomic analysis shows moderate performance in evaluating subacromial impingement syndrome (SIS) stages. This noninvasive method may aid in preliminary shoulder pain screening.
Area of Science:
- Radiology
- Machine Learning
- Musculoskeletal Imaging
Background:
- Subacromial impingement syndrome (SIS) is a common cause of shoulder pain.
- Accurate staging of SIS is crucial for effective treatment planning.
- Current diagnostic methods may have limitations in staging accuracy.
Purpose of the Study:
- To evaluate the performance of machine learning (ML)-based ultrasomic analysis for staging subacromial impingement syndrome (SIS).
- To assess the utility of ML-derived radiomic features from ultrasound for SIS staging.
Main Methods:
- Retrospective analysis of 324 SIS patients.
- Ultrasound features extracted from the supraspinatus regions.
- Machine learning models (LASSO) applied for feature selection and construction.
- Performance evaluated using Area Under the Curve (AUC) of the receiver operating characteristic.
Main Results:
- A total of 10 radiomic features were selected for model construction.
- The ultrasomics model demonstrated moderate performance in SIS stage evaluation.
- AUC values were 0.839 in the training cohort and 0.789 in the validation cohort.
Conclusions:
- Machine learning-derived ultrasomics can effectively discriminate SIS stages.
- This noninvasive, low-cost approach shows potential for preliminary screening of shoulder pain.
- Ultrasomics offers a promising tool for improving SIS diagnosis and management.
