STEP-PD: Stage-Aware and Explainable Parkinson's Disease Severity Classification Using Multimodal Clinical Assessments
Conference
Islam, MM, Templeton, JM, Poellabauer, C et al. (2026). STEP-PD: Stage-Aware and Explainable Parkinson's Disease Severity Classification Using Multimodal Clinical Assessments
. 1096-1105. 10.1109/ICHI69079.2026.00135
Islam, MM, Templeton, JM, Poellabauer, C et al. (2026). STEP-PD: Stage-Aware and Explainable Parkinson's Disease Severity Classification Using Multimodal Clinical Assessments
. 1096-1105. 10.1109/ICHI69079.2026.00135
Parkinson's disease (PD) is a progressive disorder in which symptom burden and functional impairment evolve over time, making severity staging essential for clinical monitoring and treatment planning. However, many computational studies emphasize binary PD detection and do not fully exploit repeated follow-up clinical assessments for stage-aware prediction. This study proposes STEP-PD, a severity-aware machine learning framework to classify PD severity using clinically interpretable boundaries. It leverages all available visits from the Parkinson's Progression Markers Initiative (PPMI) and integrates routinely collected subjective questionnaires and objective clinician-assessed measures. Disease severity is defined using Hoehn and Yahr staging and grouped into three clinically meaningful categories: Healthy, Mild PD (stages 1-2), and Moderate-to-Severe PD (stages 3-5). Three binary classification problems (Healthy vs. Mild, Healthy vs. Moderate-to-Severe, and Mild vs. Moderate-to-Severe) and a three-class severity task were evaluated using stratified cross-validation with imbalance-aware training. To enhance clinical interpretability, SHAP was employed to provide global explanations (including a cross-task heatmap summarizing stage-dependent symptom relevance) and local, patient-level waterfall explanations. Across all tasks, XGBoost achieved the strongest and most stable performance, with accuracies of 95.48% (Healthy vs. Mild), 99.44% (Healthy vs. Moderate-to-Severe), and 96.78% (Mild vs. Moderate-to-Severe), and 94.14% accuracy with 0.8775 Macro-F1 for three-class severity classification. Explainability results highlight a shift from early motor features (e.g., bradykinesia and tremor) to progression-related axial and balance impairments (e.g., postural instability and gait dysfunction). These findings demonstrate that multimodal clinical assessments within the PPMI cohort can support accurate and interpretable visit-level PD severity stratification.
