Paper Key : IRJ************773
Author: Vedant Sanjay Chaudhari
Date Published: 13 Oct 2023
Abstract
Parkinsons disease (PD) is a widely recognized neurological condition, the early diagnosis of which is essential for timely intervention and effective management. Recent advancements in the intersection of machine learning and speech analysis have paved the way for promising non-invasive diagnostic tools. This research leverages Support Vector Machines (SVM), a powerful machine learning algorithm, to detect PD based on speech-related features. The significance of timely PD diagnosis is underscored by the potential to improve patient outcomes. Drawing inspiration from relevant studies in the field 1-5, this research builds upon a comprehensive dataset and code framework. It employs standardization techniques for feature scaling and the SVM algorithm for predictive modeling. The primary goal is to assess the SVM model's diagnostic capabilities. Performance metrics, such as accuracy and confusion matrices, are employed to gauge the model's predictive prowess. Visualizations provide insights into data distributions, and a detailed classification report offers a holistic view of model performance. This research contributes to ongoing efforts aimed at enhancing PD diagnosis through speech analysis and machine learning. The outcomes have the potential to revolutionize diagnostic procedures, facilitating earlier intervention and more effective disease management. It represents a promising stride toward improving the quality of life for individuals affected by PD.