Related Experiment Video
Updated: Jun 29, 2025
06:28
Biomechanical Changes Related to Low Back Pain: An Innovative Tool for Movement Pattern Assessment and Treatment Evaluation in Rehabilitation
Published on: December 13, 2024
490
Comparing Video Analysis to Computerized Detection of Limb Position for the Diagnosis of Movement Control during Back
André B Peres1,2, Andrei Sancassani2, Eliane A Castro2,3
1Instituto Federal de Educação, Ciência e Tecnologia de São Paulo (IFSP), Piracicaba 13414-155, SP, Brazil.
Sensors (Basel, Switzerland)
|March 28, 2024
Summary
Computerized analysis using hidden Markov models (HMMs) can automatically detect incorrect limb positioning during weightlifting. This technology aids athletes and coaches in preventing injuries and improving performance by identifying subtle movement changes.
Area of Science:
- Biomechanics
- Sports Science
- Computer Science
Background:
- Incorrect limb positioning during weightlifting increases injury risk and compromises performance.
- Athletes and coaches often lack the expertise for self-supervision of complex lifting techniques.
- Objective, automated analysis of movement is needed to supplement human observation.
Purpose of the Study:
- To develop and validate a computerized method for detecting changes in limb position during back squat exercises.
- To assess the reliability of hidden Markov models (HMMs) in identifying movement alterations under varying loads.
- To compare the accuracy of HMMs with human expert analysis in detecting lifting technique deviations.
Main Methods:
- Hidden Markov models (HMMs) were utilized to automate the detection of joint positions and barbell trajectory.
- Ten volunteers performed back squats with 0%, 50%, and 75% body weight loads.
- Smartphone video analysis in the sagittal plane captured movement data for statistical analysis (p < 0.05).
Main Results:
- HMMs demonstrated reliability in identifying changes in movement control with added weight, showing 40% to 90% agreement with human experts.
- The study successfully trained HMMs using data from unloaded lifts to detect pattern changes.
- HMMs identified movement alterations that were imperceptible to human visual analysis.
Conclusions:
- Automated analysis using HMMs offers a reliable tool for detecting subtle changes in weightlifting technique.
- This technology can assist practitioners in self-supervising limb position and adjusting lifting techniques.
- HMMs have the potential to enhance injury prevention and performance optimization in weightlifting and resistance exercises.
