Home / Current Issue / Paper 1705496
Classification of Parkinson's using SVM
Subject area: Science,Engineering and Technology · Area of research: Information Technology
Abstract
Parkinson's disease is a neurodegenerative condition that is typified by motor symptoms, such as recognizable alterations in gait. Timely diagnosis and efficient illness management can be facilitated by the early and accurate detection of certain gait abnormalities. In this work, we combine sensor-based analysis with machine learning and deep learning approaches to offer a unique strategy for Parkinson's disease gait diagnosis. This dataset consists of 31 individuals' biological voice measures, 23 of whom have Parkinson's disease (PD). Each row in the table corresponds to one of the 195 recorded datasets from these people ("name" column), and each column in the table represents a specific frequency measure of sensors. Based on the "status" column, the primary objective of the data is to distinguish between individuals with Parkinson's disease (PD) (1) and those who are healthy (0). Our results, which show 87% accuracy using the SVM method on the dataset, show how effective this strategy is. Whether or not the person has Parkinson's, the model demonstrated robustness in identifying gait characteristics unique to the disease. Furthering our comprehension of the disease-related gait abnormalities, we also discovered critical criteria that contribute to proper categorization by examining feature importance. Our method's potential for early, non-invasive Parkinson's disease detection makes it significant. We give clinicians an objective, quantitative tool to evaluate gait disorders through the use of wearable sensors and machine learning. Early intervention and individualized treatment plans could result from this strategy in better patient care. Finally, our work presents a novel approach to gait analysis in Parkinson's disease. The outcomes highlight the method's potential to transform the way Parkinson's disease is identified using deep learning and machine learning techniques, thereby improving the lives of those who are impacted.
Keywords
Parkinson?s, Gait Abnormalities, Neurodegenerative Disorder, Machine Learning, Deep Learning, SVM
How to cite this paper
@article{1705496,
author = {Pranav Shetty, Barron Pereira, Rimsy Dua, Dr. Santosh Singh},
title = {Classification of Parkinson's using SVM},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {7},
number = {8},
pages = {171-174},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1705496.pdf},
abstract = {Parkinson's disease is a neurodegenerative condition that is typified by motor symptoms, such as recognizable alterations in gait. Timely diagnosis and efficient illness management can be facilitated by the early and accurate detection of certain gait abnormalities. In this work, we combine sensor-based analysis with machine learning and deep learning approaches to offer a unique strategy for Parkinson's disease gait diagnosis. This dataset consists of 31 individuals' biological voice measures, 23 of whom have Parkinson's disease (PD). Each row in the table corresponds to one of the 195 recorded datasets from these people ("name" column), and each column in the table represents a specific frequency measure of sensors. Based on the "status" column, the primary objective of the data is to distinguish between individuals with Parkinson's disease (PD) (1) and those who are healthy (0). Our results, which show 87% accuracy using the SVM method on the dataset, show how effective this strategy is. Whether or not the person has Parkinson's, the model demonstrated robustness in identifying gait characteristics unique to the disease. Furthering our comprehension of the disease-related gait abnormalities, we also discovered critical criteria that contribute to proper categorization by examining feature importance. Our method's potential for early, non-invasive Parkinson's disease detection makes it significant. We give clinicians an objective, quantitative tool to evaluate gait disorders through the use of wearable sensors and machine learning. Early intervention and individualized treatment plans could result from this strategy in better patient care. Finally, our work presents a novel approach to gait analysis in Parkinson's disease. The outcomes highlight the method's potential to transform the way Parkinson's disease is identified using deep learning and machine learning techniques, thereby improving the lives of those who are impacted.},
keywords = {Parkinson?s, Gait Abnormalities, Neurodegenerative Disorder, Machine Learning, Deep Learning, SVM},
month = {February},
}