Open AccessComparative Study of Regression Techniques in the Estimation of UPDRS Score for Parkinson’s disease
Author(s) -
B. Santhi,
Harini Ram Prasad,
Rohith Jayaraman
Publication year - 2018
Publication title -
international journal of engineering and technology
Language(s) - English
Resource type - Journals
ISSN - 2227-524X
DOI - 10.14419/ijet.v7i3.12.16498
Subject(s) - rating scale , parkinson's disease , regression , disease , multilinear map , regression analysis , metric (unit) , physical therapy , lasso (programming language) , psychology , physical medicine and rehabilitation , medicine , statistics , computer science , mathematics , engineering , operations management , pure mathematics , world wide web
Studies have shown that instances of Parkinson’s disease have been on the rise over the past 30 years. A metric that measures the extremity of Parkinson’s disease in a person is their Unified Parkinson’s Disease Rating Scale (UPDRS) score. Thus, an algorithm that can predict the UPDRS score of a Parkinson’s patient will be effective in determining the severity of the patient’s condition. This paper aims to forecast a patient’s UPDRS score by inferring patterns from historical figures and other independent parameter values that affect the patients’ UPDRS score. Four regression techniques namely multilinear, ridge, robust and LASSO regression are being used to predict the UPDRS scores. This will be done using the R language and through the use of the MASS, glmnet packages.