First, the investigators aim to develop an artificial intelligence algorithm for the identification and quantification of FOG episodes using Red-Green-Blue (RGB) and walkway pressure data. PD patients experiencing FOG will undergo a standardized walking protocol on the PKMAS System, acquiring simultaneous data from two angles of video recording and foot pressure. Labeled videos of FOG incidents will be inputted into an algorithm, Adaptive Fusion Algorithm for Spam Detection (AFSD), utilizing raw RGB data, optical flow data (DualTVL), and their combination for temporal segmentation.
Second, the investigators intend to pinpoint sources of FOG in PD patients through electrophysiological methods during unconstrained walking, employing a high-density 64-channel electroencephalographic ambulatory recording and motor imagery fMRI to delve into the pathophysiology of FOG under different ambulatory conditions. Effective connectivity among higher neural regions, including the basal ganglia, cerebellum, and cortical regions, will be explored. The resulting connectivity map will be overlaid with electric recording data to examine neurovascular coupling or uncoupling.
Third, the investigators plan to conduct non-invasive interventions (transcranial direct current stimulation: tDCS, magnetic resonance-guided focused ultrasound: MRgFUS) and Deep Brain Stimulation (DBS) to assess their impact on gait and FOG in PD patients. Motor imagery fMRI and electrophysiological investigations will be performed before and after treatment using the mentioned methods. Our multi-modal approach aims to unravel the complexities of FOG in PD patients, providing valuable insights and potentially benefiting those enduring the challenges posed by this debilitating symptom.