Colloquium: Koen den Hertog (C&O)

30 August 2022 09:30 - Location: Meeting Room 1, FACULTY OF AEROSPACE ENGINEERING, KLUYVERWEG 1, DELFT | Add to my calendar

Monitoring symptom progression in Parkinson’s Disease using Least Squares Support Vector Regression

Parkinson’s Disease is a neurodegenerative disease, that has a decline in motor behaviour as one of its symptoms. This decline is currently monitored using subjective measures, such as questionnaires. Quantification of this decline can improve treatment and allow for earlier identification of the disease. Earlier work in this area lead to the development of a proof-of-concept for detecting behavioural changes in motor performance, by using linear regression techniques and data obtained from a cybernetic tracking task, in which participants were asked to perform the task in a series of trials taking place over multiple days. This paper improves upon that work by applying a machine learning model to simulated data based on the measured data, with the aim of getting more detections and earlier detection of changes. The machine learning model used in this paper is Least Squares Support Vector Regression, and the results are promising. Depending on the model settings an improvement of 80% in the amount of detections was reached, and detection of changes can be sped up by 30%. In addition, the influence of the different hyperparameters was investigated, as well as making use of predictive modelling. Overall, the new model shows definite improvement over the earlier model, and can be developed further in subsequent steps towards a clinical applicable method.

Supervisor: Daan Pool

/* */