Identifying Behavioural Changes due to Parkinson’s Disease Progression in Motor Performance Data
Development of a Tool for Monitoring Treatment of Parkinson’s Disease
L.A. Lugtenborg (TU Delft - Aerospace Engineering)
Daan M. Pool – Graduation committee member (TU Delft - Control & Simulation)
M Mulder – Mentor (TU Delft - Control & Simulation)
Johan J.M. Pel – Graduation committee member (Erasmus MC)
More Info
expand_more
Other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as Creative Commons.
Abstract
Parkinson's disease can severely affect motor performance and impede in executing daily activities. Treatment can greatly improve patients' quality-of-life, however, disease detection and monitoring is still performed subjectively. Quantification of patients' motor performance and its decline due to increasing symptom severity using tracking tasks could provide a solution and even help in early disease assessment. In order to develop a proof-of-concept for a tool that can be used for the detection of behavioural changes in motor performance data, the longitudinal clinical data are approximated by a combined data-set with experimental data of healthy participants and simulated Parkinson's disease control behaviour. 25 healthy participants in the age range of 55-75 participated in a manual pursuit tracking experiment to identify baseline control behaviour. PD data were simulated by bootstrapping the experimental data and scaling this value based on previous research. The resulting experimental and PD data were combined and a general linear regression model was used to see if a change in control behaviour due to upcoming PD symptoms could be detected with trend analysis. It was found that for the parameters related to a decline in motor performance caused by the disease, for at least 50% of the participants a simulated change in motor behaviour was successfully detected. This means that the developed method is able to detect a trend for half of the population and is a major step forward in the development of a tool that can aid monitoring of disease progression and treatment for Parkinson's disease.