Details

Autor(en) / Beteiligte

• Olatunji, Sunday O.
• Alzahrani, Abdulaziz
• Alsubaie, Abdulaziz
• Aldahlan, Osamah
• Alnafea, Ibrahim
• Alamri, Abdullah
• Basheer Ahmed, Mohammed Imran
• Alam Khan, Mohammad Aftab
• Almuhaidib, Abdulrahman
• Alhiyafi, Jamal

Titel

Machine Learning Based Preemptive Diagnosis of Parkinson's Disease Using Saudi Clinical Data: A Preliminary Case Study on Saudi Arabia Dataset

Ist Teil von

• 2023 Fourth International Conference on Intelligent Data Science Technologies and Applications (IDSTA), 2023, p.1-6

Ort / Verlag

IEEE

Erscheinungsjahr

2023

Link zum Volltext

Quelle

IEL

Beschreibungen/Notizen

• Parkinson's disease is a degenerative neurological condition that gradually worsens over time and currently has no known cure. It affects the nervous system and can cause thinking difficulties, loss of automatic movements, and problems with balance. Given its impact on societies and individuals alike, detecting Parkinson's disease early can help reduce the incidence and severity of the condition. Therefore, this study aims to predict Parkinson's disease preemptively to avoid complications from the disease. The study employed two machine learning algorithms, Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN), to predict the disease. GridSearchCV with 10-folds was utilized to improve the accuracy of the models. In addition, Sequential Forward Feature Selection (SFFS) was employed to reduce the feature number and increase accuracy. The results showed that, using 27 features, K-NN performed best in terms of accuracy (81%), recall (81%), and precision (82%).