Data Analysis on Parkinson’s Biomarkers: A study on Mean of Putamen, Mean of Caudate, Total UPDRS Score, and Age Onset of PD

Date
2021-01-01
Authors
Malepati, Anudeep
Major Professor
Dr. Sree Nilakanta
Advisor
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Journal Issue
Series
Department
Information Systems and Business Analytics
Abstract

Parkinson’s disease is the second-most common neurodegenerative disorder in the United States. Alzheimer’s and Parkinson’s disease are the most common neurodegenerative diseases world-wide. This study is carried out to evaluate four biomarkers which are significant in assessing the progression of Parkinson’s Disease (PD). Like all neurodegenerative diseases, the symptoms of the Parkinson’s Disease do not manifest until the disease progression, this leads to importance of identifying changes to the biomarkers significant to the progression of disease. Biomarkers from files like Biospecimen results, Unified Parkinson’ Disease Rating Scale (UPDRS), Non-motor skill tests, Imaging tests, and patient’s medical history are used to assess their influence the four proposed biomarkers.

A significant amount of research went in to analyzing various biomarkers to understand Parkinson’s disease progression. The change in the biomarkers has changed the disease outcome. This study takes a different path in analyzing the biomarkers in the PPMI database. A good amount of research shows how some biomarkers influence the disease outcome more than other biomarkers in PPMI. So, studying these most influential biomarkers change with the change outcome of the other biomarkers. The data for this study is collected from Parkinson’s Progression Markers Initiative (PPMI) an observational study on Parkinson’s disease. This study addresses data munging, manipulation, cleaning, imputing missing values, univariate analysis along with employing commonly used regression Machine Learning (ML) models, and discusses the results of the ML models. Study concludes by finding the most significant features which influenced in predicting the outcome of the models.

Comments
Description
Keywords
Citation
DOI
Source