Unveiling Predictors of Parkinson's Disease Progression: A Machine Learning Approach with Questionnaire and Wearable...

Area of Science:

• Neurology
• Biomedical Engineering
• Data Science

Background:

• Parkinson's disease (PD) is a progressive neurodegenerative disorder impacting motor and non-motor functions, significantly reducing patient quality of life.
• Accurate prognostic markers are crucial for effective management and personalized treatment strategies in PD.
• Current monitoring methods may not fully capture the nuances of disease progression.

Purpose of the Study:

• To employ machine learning (ML) for identifying prognostic markers of Parkinson's disease progression.
• To integrate diverse data sources, including questionnaires and wearable monitoring devices, for enhanced predictive accuracy.
• To leverage feature engineering on wearable data to uncover novel prognostic indicators.

Main Methods:

• Longitudinal study design analyzing patient data over time.
• Application of machine learning algorithms to identify significant prognostic features.
• Integration of data from the Unified Parkinson's Disease Rating Scale (UPDRS) Part III questionnaire and wearable sensor data.
• Utilizing feature engineering techniques to extract unique insights from wearable device data.

Main Results:

• The UPDRS Part III questionnaire was identified as the most statistically significant factor correlated with PD progression.
• Engineered features derived from wearable devices demonstrated higher importance scores compared to other features.
• The combination of ML with diverse data sources provided valuable insights into disease progression dynamics.

Conclusions:

• Machine learning, combined with patient-reported outcomes and wearable sensor data, offers a powerful approach to understanding Parkinson's disease progression.
• The UPDRS Part III and engineered wearable features are key indicators for monitoring PD advancement.
• These findings can guide personalized therapeutic interventions and improve clinical monitoring of Parkinson's disease.