Interpretable Classification of Early Stage Parkinson’s Disease from EEG

Detecting early stage Parkinson’s Disease using electroencephalogram (EEG) data presents a significant challenge. This paper introduces a novel approach, representing EEG data as a 15-variate series of bandpower and peak frequency values/coefficients. We hypothesise that this representation captures essential information from the noisy EEG signal, improving disease detection. Statistical features extracted from this representation are input to interpretable machine learning models, specifically Decision Tree and AdaBoost classifiers. Our pipeline is deployed within our proposed framework which enables high-importance data types and brain regions for classification to be identified. Our analysis reveals that while there is no significant regional importance, the N1 sleep data type exhibits statistically significant predictive power (p<0.01) for early-stage Parkinson’s. AdaBoost classifiers trained on N1 data consistently outperform baseline models, achieving over 80% accuracy and recall. Our classification pipeline statistically significantly outperforms baseline models. Paired with the interpretability of our pipeline this enables us to generate meaningful insights into the classification of early stage Parkinson’s.