Planting Synchronisation Trees for Discovering Interaction Patterns among Brain Regions
- Author(s)
- Lena Bauer, Philipp Grohs, Afra Wohlschläger, Claudia Plant
- Abstract
The proposed data mining method is designed to analyse the synchronisation behaviour of multiple time series with the Kuramoto model which we use to construct synchronisation trees. By transforming time series data with the Hilbert transform, the initial phases of multiple time series can be provided to the model and subsequently the synchronisation process is represented by a tree structure, which can then further be analysed, e.g., by comparing tree edit distances. The proposed analysis might be interesting in the context of neuroscience as brain activity of a subject is often represented by time series corresponding to different brain regions. Discovering certain synchronisation patterns is then useful, when alterations of those patterns can be observed in different pathologies or brain states.
- Organisation(s)
- Research Network Data Science, Department of Mathematics, Research Group Data Mining and Machine Learning
- External organisation(s)
- Technische Universität München
- Pages
- 1035-1036
- No. of pages
- 2
- DOI
- https://doi.org/10.1109/ICDMW.2019.00149
- Publication date
- 11-2019
- Peer reviewed
- Yes
- Austrian Fields of Science 2012
- 102033 Data mining
- Keywords
- ASJC Scopus subject areas
- Software, Computer Science Applications
- Portal url
- https://ucris.univie.ac.at/portal/en/publications/planting-synchronisation-trees-for-discovering-interaction-patterns-among-brain-regions(45e76753-bedf-4a54-bf4e-b5ab4f8305d6).html