Loading [MathJax]/extensions/MathZoom.js
An Intelligent Technique for the Effective Prediction of Parkinson Disease | IEEE Conference Publication | IEEE Xplore

An Intelligent Technique for the Effective Prediction of Parkinson Disease


Abstract:

Parkinson is a condition of the nervous system that is linked to breakdown of basal ganglia of the brain. This disorder is responsible for affecting the neurological syst...Show More

Abstract:

Parkinson is a condition of the nervous system that is linked to breakdown of basal ganglia of the brain. This disorder is responsible for affecting the neurological system and other body parts that are nerve-controlled. Its symptoms vary from person to person, however common symptoms include Tremor, Slow Movement and Stiffness in Muscles, Impaired Posture, Speech Changes and Loss of Automatic Movements. The main cause of this disease is not known, that is why it is not curable and has no proven prevention yet. This research concerns the application of machine learning classifiers including KNN, Logistic Regression, Decision Tree, Naive Bayes and Random Forest for the detection of Parkinson's disease. We are using a Parkinson disease dataset using a machine learning algorithm for detection to determine the performance in terms of accuracy. In a nut shell, the Random Forest classifier, without using SMOTE, provided the highest accuracy among the models considered. However, this result and existing results as well are based on imbalanced class. This research is contributing valuably for the convergence of machine learning and healthcare for effective predictions of this disease using Smote for balancing out same dataset classes with accuracy of 98.7%. The results hold significance for future modifications in medical diagnostics for more accurate and effective detection methodologies.
Date of Conference: 23-25 July 2024
Date Added to IEEE Xplore: 04 December 2024
ISBN Information:
Conference Location: Windhoek, Namibia

References

References is not available for this document.