Can wearable sensor based measures of gait accurately reflect Parkinson's disease severity? A systematic review and meta-analysis

Abstract

Background: Classical clinical assessments of Parkinson's disease (PD) severity are subjective and prone to variability. Wearable sensors, particularly inertial measurement units (IMUs), offer a quantitative approach to assessing gait abnormalities, which are critical markers of disease progression. Objectives: Herein, we aimed to evaluate the ability of IMUs in detecting the severity of motor symptoms in PD by examining the relationship between IMU-derived gait parameters and established disease severity scales, including the Movement Disorder Society-Unified PD Rating Scale (MDS-UPDRS III), Hoehn and Yahr (H&Y) scale, and Postural Instability and Gait Disorders (PIGD) scale. Methods: A systematic search was conducted across PubMed, Web of Science, and Cochrane Central databases. Studies were included if they assessed gait using IMUs and reported correlations with PD scales. Data extraction included study characteristics, participant demographics, sensor details, and gait assessment protocols. Results: The review included 93 studies, with 21 providing data for meta-analysis. Key findings indicated that slower gait velocity was associated with greater symptom severity, as measured by MDS-UPDRS III and H&Y. Severe PD is charactrized by shorter step-length, longer stance time and slower turning-velocity, as measured by MDS-UPDRS III. Advanced H&Y stages were associated with reduced stride-length, toe-off, and strike angles. Sensor distribution analysis showed a focus on lower body kinematics, with the feet and waist being the most common sensor locations. Conclusions: Gait velocity, stance time, and strike angle, show promise as objective digital biomarkers for assessing PD severity. These findings highlight the potential of IMU to complement traditional clinical assessments. © 2025 Elsevier Ltd