Toward future diagnostics of Parkinson's disease: a perspective on multimodal motor assessment and personalized...

The accurate diagnosis of Parkinson's disease (PD) remains challenging due to the heterogeneity of motor symptoms and the reliance on subjective clinical assessments. Although biomarkers such as dopaminergic imaging, fluid assays, and genetic testing can support diagnosis, their clinical utility is limited by cost, accessibility, and low sensitivity to early motor changes. Consequently, diagnosis depends largely on clinical observation, highlighting the need for more accurate, objective, and integrable tools to detect early motor symptoms. Advances in wearable sensing and computational modeling offer opportunities to address this gap. Multimodal approaches combining inertial measurement units (IMUs) and surface electromyography provide complementary information on movement and neuromuscular control. Moreover, IMU-derived kinematics can drive personalized musculoskeletal simulations, generating digital twins that estimate internal biomechanical quantities not directly observable from wearable data alone. This perspective aims to synthesize evidence and outline a clinically grounded framework for multimodal data fusion in PD, in which wearable- and model-derived features can be jointly analyzed. This approach enables the construction of a feature space capturing kinematic, neuromuscular, and biomechanical aspects of motor dysfunction, from which discriminative digital biomarkers may be derived. Rather than proposing new sensing or modeling technologies, the novelty lies in advancing a targeted integration strategy that evaluates which wearable modalities-and combinations-are most suitable for quantifying PD and driving musculoskeletal digital twins. Importantly, the framework is grounded in a protocol co-designed with movement-disorder specialists, ensuring clinically meaningful task selection, workflow compatibility, feasibility, and patient acceptability from the outset, supporting more interpretable and personalized diagnostic strategies for PD.