Predicting Improvement in Parkinson's Disease Patients using Location of Stimulation

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) has become a standard therapy for Parkinson's disease (PD). However, the capability to predict symptom improvement based on specific electrode locations has not yet been developed. In this study, we analyzed pre- and postoperative MRI/CT data and stimulation parameters from PD patients to map electrode locations and estimate the volume of electric field (VEF) around the STN. We then modeled how these features were related to improvements in PD symptoms, measured using the Unified Parkinson's Disease Rating Scale (UPDRS-III). Our predictive models accurately predicted improvement scores for total UPDRS scores (MSE = 13.92, R² = 0.78) and individual symptoms such as rigidity (MSE = 0.107, R² = 0.98), tremor (MSE = 1.56, R² = 0.85), bradykinesia (MSE = 4.67, R² = 0.85), and freezing of gait (MSE = 0.28, R² = 0.84). These findings support the development of models for symptom-specific targeting in DBS, with the STN associative VEF shown to be a key predictive feature, potentially improving clinical outcomes and patient quality of life.Clinical relevance-This study provides predictive models that help clinicians tailor DBS targeting to individual Parkinson's symptom profiles based on stimulation location.