Editorial: Advancing the treatment landscape in Parkinson's disease using sensor technology and data-driven modeling

Effectively managing Parkinson's disease (PD) symptoms is a formidable challenge for healthcare providers. PD patients exhibit considerable clinical heterogeneity in their symptoms, yet many patients struggle with receiving specialized care to personalize their treatment strategies. Advanced PD is associated with emergence of motor complications (i.e., motor fluctuations and dyskinesia), axial symptoms including gait, postural and balance disorder as well as freezing of gait (FoG), and non-motor symptoms such as sleep disturbances and neurobehavioral disorders. Studies classifying Parkinson's clinical subtypes have predominantly relied on clinimetrics. Data driven approaches incorporating motor and non-motor symptoms associated with the disease have expanded the clinical spectrum with recent studies demonstrating overarching phenotypic clusters: mild motor predominant, non-motor predominant, intermediate, diffuse malignant (Fereshtehnejad et al., 2017; Mu et al., 2017). However, these phenotypic classifications and treatment response measurements remain confined to non-granular, subjective longitudinal assessments.