All projects
Machine Learning / Healthcare
Machine Learning Approaches for Early Detection of Parkinson's Disease
Machine learning models for early Parkinson's disease detection from voice features.
Method
SMOTE for class imbalance, patient-level splitting, and SHAP interpretability.
Results
90%
Accuracy
93.33%
F1-score
XGBoost
Best model