Gait analysis and prediction in Parkinson's disease using rhythmic auditory stimulation: A data-driven approach with...

Area of Science:

• Neurology
• Biomedical Engineering
• Rehabilitation Science

Background:

• Gait disturbance is a primary Parkinson's disease (PD) symptom, leading to reduced mobility and increased fall risk.
• Rhythmic auditory stimulation (RAS) is a potential intervention to improve PD gait, but patient responses vary.
• Predicting RAS responders is crucial for developing personalized treatment strategies.

Purpose of the Study:

• To evaluate the effects of Rhythmic Auditory Stimulation (RAS) on gait parameters in Parkinson's disease (PD) patients.
• To develop and validate a machine learning model for predicting individual patient responses to RAS.
• To identify key clinical and gait features that predict responsiveness to RAS.

Main Methods:

• Three hundred PD patients were enrolled in a randomized controlled trial with RAS or control conditions.
• Gait analysis utilized 3D motion capture and force plates, with baseline assessments including UPDRS-III and MoCA scores.
• Machine learning models (Random Forest, XGBoost, SVM) were trained using demographic and baseline gait data to predict RAS responsiveness.

Main Results:

• Rhythmic Auditory Stimulation (RAS) significantly improved gait speed, stride length, and reduced double support time and variability in Parkinson's disease (PD) patients (p < 0.001).
• Approximately 67.8% of patients in the RAS group were identified as responders, showing improvements in gait speed and double support time.
• The Random Forest model achieved the best predictive performance (AUC=0.713), with UPDRS-III score, stride length, MoCA score, and gait asymmetry identified as key predictors.

Conclusions:

• Rhythmic Auditory Stimulation (RAS) is an effective intervention for enhancing gait performance in Parkinson's disease (PD) patients.
• A machine learning framework can reasonably predict RAS responders, facilitating personalized gait rehabilitation strategies.
• Identifying key predictors like UPDRS-III and gait parameters allows for tailored therapeutic approaches in PD management.