الجهة البحثية: جامعة البترا
عنوان البحث المنشور:
A CatBoost Predictive Model for Parkinson's Disease Early Detection
سنة النشر: 2025
This study explores the use of the CatBoost classifier on the UCI Parkinson's dataset, emphasising its effectiveness in distinguishing between healthy individuals and those affected by Parkinson's disease (PD). The CatBoost model demonstrated remarkable accuracy, achieving 100% for both classes and effectively classifying every instance in the dataset. Moreover, the model showed flawless precision and recall, measured at 1.00 (or 100%), signifying the absence of false positives or overlooked cases in its predictions. The F1-score, which balances precision and recall, also achieved a perfect score 1.00 for both classes, highlighting the model's outstanding performance. The validation dataset included 30 healthy individuals and 29 patients with Parkinson's disease, showcasing the model's ability to utilise voice features for precise diagnosis and monitoring of the condition. The results underscore the effectiveness of the CatBoost classifier as a reliable instrument for the early identification and management of neurodegenerative disorders.
رابط البحث المنشور
https://ieeexplore.ieee.org/abstract/document/11013735