SCOPE-PD: Explainable AI on Subjective and Clinical Objective Measurements of Parkinson's Disease for Precision Decision-Making

High accuracy with combined data: The SCOPE-PD framework achieved 98.66% accuracy in predicting Parkinson's disease by integrating both subjective patient reports and objective clinical measurements, demonstrating that combining multiple data types produces better results than relying on subjective assessments alone.

Identifies key physical indicators: The research pinpointed tremor, bradykinesia (slow movement), and facial expression changes as the three most important physical signs for predicting PD, providing clinicians with specific markers to prioritize during evaluation.

Addresses diagnostic delays through explainability: By using explainable AI (SHAP-based analysis), the framework makes its predictions transparent and interpretable, helping doctors understand why the system flagged a patient as at-risk rather than treating it as a "black box"—a critical feature for building trust in AI-assisted diagnosis.

Tackles subjectivity in diagnosis: Traditional PD diagnosis relies heavily on subjective clinical judgment, which can delay diagnosis; this framework reduces diagnostic uncertainty by incorporating objective measurements alongside subjective data to support more consistent, earlier identification.

Enables personalized decision-making: Rather than applying one-size-fits-all diagnostic criteria, SCOPE-PD provides individualized risk estimates for each patient, supporting precision medicine approaches that could lead to more tailored treatment plans.