NeuroTrace: An iPad-Based Handwriting Analysis System for Parkinson's Disease Screening

NeuroTrace is a research prototype for exploring digital handwriting biomarkers associated with Parkinson’s disease using an iPad and Apple Pencil. The system collects fine-grained handwriting and drawing data from tasks including spiral drawing, static pen control, tapping, sentence writing, wave drawing, circle drawing, and clock drawing.

The project investigates whether temporal and spatial handwriting features can be used for machine-learning-based screening of Parkinsonian motor characteristics. Data processing includes temporal resampling and feature extraction, followed by classification using machine learning models including Random Forest and XGBoost. Model evaluation is designed around subject-independent validation such as leave-one-subject-out cross-validation.

NeuroTrace is intended as an experimental research and educational system for studying digital biomarkers and human-computer interaction in neurological screening. It is not a clinical diagnostic tool and should not be used as a substitute for professional medical evaluation.

This release contains the software, model-related code, documentation, and associated research materials for the NeuroTrace project.