Measurement of Shoulder Abduction Angle with Posture Estimation Artificial Intelligence Model

Abstract

Substantial advancements in markerless motion capture accuracy exist, but discrepancies persist when measuring joint angles compared to those taken with a goniometer. This study integrates machine learning techniques with markerless motion capture, with an aim to enhance this accuracy. Two artificial intelligence-based libraries—MediaPipe and LightGBM—were employed in executing markerless motion capture and shoulder abduction angle estimation. The motion of ten healthy volunteers was captured using smartphone cameras with right shoulder abduction angles ranging from 10° to 160°. The cameras were set diagonally at 45°, 30°, 15°, 0°, −15°, or −30° relative to the participant situated at a distance of 3 m. To estimate the abduction angle, machine learning models were developed considering the angle data from the goniometer as the ground truth. The model performance was evaluated using the coefficient of determination R2 and mean absolute percentage error, which were 0.988 and 1.539%, respectively, for the trained model. This approach could estimate the shoulder abduction angle, even if the camera was positioned diagonally with respect to the object. Thus, the proposed models can be utilized for the real-time estimation of shoulder motion during rehabilitation or sports motion.

Estimating the Thumb Rotation Angle by Using a Tablet Device With a Posture Estimation Artificial Intelligence Model

“…According to one study, the integration of MediaPipe with artificial intelligence (AI) models for posture analysis can improve accuracy in assessing shoulder abduction angles [20]. In this study, the mean absolute…”

“…Villalba-Meneses et al described the classification using ML models such as SVM of motion capture data obtained from the range of motion exercises among healthy and clinically diagnosed patients with low back pain [27]. Kusunose et al indicated that the integration of MediaPipe with ML models using liner regression and LightGBM for posture analysis has improved accuracy in assessing shoulder abduction angles [20]. In this study, four ML algorithms -linear regression, ElasticNet, SVM, and LightGBMwere applied and the LightGBM model demonstrated a high accuracy, as the change in each parameter relative to the true value of each parameter indicated a nonlinear change.…”

“…In the field of AI, the explainability of ML models is important. Prior research has incorporated feature importance and SHAP analyses to ensure transparency and interpretability [20,22]. In the present study, the feature values that were important for predicting the rotation angle (θ) were also assessed utilizing feature importance and the SHAP values.…”