Dysphonia measures in parkinson's disease and their use in prediction of its progression

Kambhampati, Chandrasekhar; Sarangdhar, Mayur; Poolsawad, Nongnuch

Abstract

Parkinson's Disease (PD) is a neurodegenerative disorder that impairs the motor skills, speech and general muscle coordination. The progression of PD is assessed using a clinically defined rating scale known as Unified Parkinson's Disease Rating Scale (UPDRS). Recent studies have shown the use of telemonitoring of PD using simple speech tests which replicate the UPDRS to clinician's accuracy. Regression analysis is performed on a database of speech recordings of 42 PD patients to analyse the relation between dysphonia measures and the UPDRS with the progression of PD. It is observed that there is a strong correlation between the dysphonia measures and the UPDRS and it is possible to predict the UPDRS scores weekly using linear regression techniques. The results also suggest that certain dysphonia measures evolve more significantly with the progression in PD. This is supported by Principle Component Analysis (PCA) which identifies the dysphonia measures that are strongly correlated during the course of PD progression. The data is classed by trials undertaken by the patients and each patient had at least 20 valid trials.

Citation

Kambhampati, C., Sarangdhar, M., & Poolsawad, N. (2010). Dysphonia measures in parkinson's disease and their use in prediction of its progression.

• Conference Name: International Conference on Knowledge Engineering and Ontology Development
• Conference Location: Valencia, Spain
• Start Date: Oct 25, 2010
• End Date: Oct 28, 2010
• Acceptance Date: Dec 31, 2010
• Publication Date: Dec 31, 2010
• Journal: KEOD 2010 - Proceedings of the International Conference on Knowledge Engineering and Ontology Development
• Pages: 104 - 109
• Series ISSN: 978-989-8425-29-4
• ISBN: 978-989-8425-29-4
• Public URL: https://hull-repository.worktribe.com/output/409677