University of Twente Student Theses
The correlation between outcome measures of smartwatch data and MDS-UPDRS part-III scores in parkinson’s disease patients
Berendsen, Stef Hendrikus Godefridus and Meuwese, Job and Schoenmakers, Loet (2023) The correlation between outcome measures of smartwatch data and MDS-UPDRS part-III scores in parkinson’s disease patients.
|
PDF
4MB |
| Abstract: | Parkinson’s disease (PD) is the second most common neurodegenerative disease, and the number of Parkinson’s patients is expected to double in the coming years [1]. The increase in the number of patients could cause problems with the current treatment method. Parkinson’s disease cannot be cured, the symptoms can only be reduced. This can be done through medication or deep brain stimulation (DBS). The Amsterdam UMC is working on a study that should ensure that deep brain stimulation becomes adaptive (aDBS). So that the patient and/or doctor can immediately change the strength and amount of DBS. The study uses smartwatches to detect tremor and dyskinesia and an MDS-UPDRS scoresheet to consider current status. This research provides a first set-up for processing and analysing the smartwatch and MDS-UPDRS data. In addition, research has been conducted into whether there is a correlation between the smartwatch data and the MDS-UPDRS score list. This eventually led to the following research question: What is the correlation between outcome measures of the smartwatch data and MDS-UPDRS scores of patients with Parkinson’s disease? With the help of a literature search, the researchers have formed a basis about Parkinson’s disease. But also about current techniques such as machine learning and artificial intelligence to see whether they can be combined with current and future treatment methods. After this, scripts were written using Python to process, analyse and correlate the available data from smartwatches and MDS-UPDRS. Using 3 iterations of a Random Forest (RF) algorithm, the RF had average accuracies ranging from 48,9% to 52,7%. Accuracies for values within one of the correct value ranged from 80,9% to 84,8%. Next to this, data on probabilities of tremor and dyskinesia, MDS-UPDRS Part-III scores and time availability were plotted. With this retrieved data and algorithms, we concluded there to not be a correlation between smartwatch data and MDS-UPDRS Part-III scores. |
| Item Type: | Essay (Bachelor) |
| Clients: | Amsterdam UMC, Amsterdam, Netherlands |
| Faculty: | TNW: Science and Technology |
| Subject: | 44 medicine |
| Programme: | Technical Medicine BSc (50033) |
| Link to this item: | https://purl.utwente.nl/essays/95209 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page