Abstract:Background: The accuracy of human pose tracking using smartphone camera (2D-pose) to quantify shoulder range of motion (RoM) is not determined. Methods: Twenty healthy individuals were recruited and performed shoulder abduction, adduction, flexion, or extension, captured simultaneously using a smartphone-based human pose estimation algorithm (Apple's vision framework) and using a skin marker-based 3D motion capture system. Validity was assessed by comparing the 2D-pose outcomes against a well-established 3D motion capture protocol. In addition, the impact of iPhone positioning was investigated using three smartphones in multiple vertical and horizontal positions. The relationship and validity were analysed using linear mixed models and Bland-Altman analysis. Results: We found that 2D-pose-based shoulder RoM was consistent with 3D motion capture (linear mixed model: R2 > 0.93) but was somewhat overestimated by the smartphone. Differences were dependent on shoulder movement type and RoM amplitude, with adduction the worst performer among all tested movements. All motion types were described using linear equations. Correction methods are provided to correct potential out-of-plane shoulder movements. Conclusions: Shoulder RoM estimated using a smartphone camera is consistent with 3D motion-capture-derived RoM; however, differences between the systems were observed and are likely explained by differences in thoracic frame definitions.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to evaluate the accuracy of using smartphone cameras for human pose estimation (2D - pose) to quantify the range of motion (RoM) of the shoulder joint, and compare it with the traditional 3D motion capture system (3D - Motion Capture, 3D - Mocap). Specifically, the main objectives of the study include: 1. **Determine the accuracy and effectiveness of 2D - pose in estimating the range of motion of the shoulder joint**: By comparing the 2D - pose estimation results based on the Apple Vision framework with the results of the 3D motion capture system using skin markers, verify the accuracy of the former in quantifying the range of motion of the shoulder joint. 2. **Demonstrate and provide methods to compensate for potential out - of - plane motions**: The study explored the influence of different mobile phone placement positions (vertical and horizontal directions) on the measurement results, and proposed correction methods to reduce the errors caused by these factors. ### Background and Significance Shoulder joint pain is an important cause of disability and affects the function of the shoulder joint. Whether it is conservative treatment or surgical treatment, regularly tracking the function of the shoulder joint is crucial for evaluating the effectiveness of interventions. Current clinical methods such as goniometers and visual estimation have deficiencies in accuracy and consistency, so a more objective and easy - to - use evaluation tool is needed. Smartphone video imaging technology (2D - pose) as an emerging tool has the advantage of being simple and easy to use in outpatient clinics or at home, but its accuracy needs further verification. ### Research Methods - **Participants**: Twenty healthy volunteers were recruited to perform shoulder abduction, adduction, flexion, and extension movements. - **Experimental Setup**: Twelve Vicon cameras (3D - Mocap system) and two iPhone 13s and one iPhone 13 Pro (running the mymobility®App) were used to simultaneously capture shoulder joint movements. - **Data Analysis**: The results of 2D - pose and 3D - Mocap were compared through linear mixed models and Bland - Altman analysis to evaluate their correlation and consistency. ### Main Findings - **Consistency**: The range of motion of the shoulder joint estimated by 2D - pose is highly consistent with the 3D - Mocap results (R² > 0.93), but there is a certain overestimation phenomenon, especially in the medium - amplitude range of motion. - **Influencing Factors**: Different types of shoulder joint movements and RoM amplitudes have a significant impact on the results, among which the adduction movement has the worst performance. - **Correction Methods**: A correction method for potential out - of - plane motions is provided to improve the accuracy of measurement. ### Conclusion Although the range of motion of the shoulder joint estimated by 2D - pose is basically consistent with the 3D - Mocap results, there are still differences between the two, which may be due to the different definitions of the thoracic reference frame. Future research can further optimize the 2D - pose algorithm to reduce these differences and make it a reliable clinical evaluation tool.

Comparison of Shoulder Range of Motion Quantified with Mobile Phone Video-Based Skeletal Tracking and 3D Motion Capture—Preliminary Study

Accuracy of a Smartphone Software Application Compared With a Handheld Goniometer for Measuring Shoulder Range of Motion in Asymptomatic Adults

How do digital range of motion measurement devices 'measure-up' to traditional goniometry in assessing shoulder range of motion? A systematic review and meta-analysis

Estimation of Shoulder Joint Rotation Angle Using Tablet Device and Pose Estimation Artificial Intelligence Model

REAL-TIME HIP RANGE OF MOVEMENT MEASUREMENT IN TELEREHABILITATION; COMPARISON OF HUMAN POSE ESTIMATION TOOL, PHYSIOROM, VS 3D MOTION CAPTURE SYSTEM

Algorithmic assessment of shoulder function using smartphone video capture and machine learning

A Photogrammetric Method for the Measurement of Three-Dimensional Cervical Range of Motion

A Single Video Camera Postural Assessment System to Measure Rotation of the Shoulder During Computer Use

Estimation of center of pressure information by smartphone sensors for postural control training

Comparative accuracy of a shoulder range motion measurement sensor and Vicon 3D motion capture for shoulder abduction in frozen shoulder

Assessing Spatiotemporal Behavior of Human Gait: A Comparative Study Between Low-Cost Smartphone-Based Mocap and OptiTrack Systems

Development and Validation of a Mobile Phone Application for Measuring Knee Range of Motion

Smartphone videos-driven musculoskeletal multibody dynamics modelling workflow to estimate the lower limb joint contact forces and ground reaction forces

Smartphone three-dimensional imaging for body composition assessment using non-rigid avatar reconstruction

Effective evaluation of HGcnMLP method for markerless 3D pose estimation of musculoskeletal diseases patients based on smartphone monocular video

Accuracy of a computer vision system for estimating biomechanical measures of body function in axial spondyloarthropathy patients and healthy subjects

Is Markerless More or Less? Comparing a Smartphone Computer Vision Method for Equine Lameness Assessment to Multi-Camera Motion Capture

Reliability of different smartphones measuring the hallux valgus parameters in a new rapid method: a follow-up study

Feasibility, reliability and validity of self-measurement of knee range-of-motion using an accelerometer-based smartphone application by patients with total knee arthroplasty

Agreement between a 3D camera system and an inertial measurement unit for assessing the range of motion, head repositioning accuracy and quality of movement during neck and head movements

The Validity of a Smartphone-Based Method for Acquiring 3D Images of the Face