|
Activity Number:
|
536
- Contributed Poster Presentations: Section on Statistics in Imaging
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 1, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics in Imaging
|
|
Abstract #329656
|
|
|
Title:
|
Switching Regimes Time Series Models with Application to Changes Brain Connectivity in an fMRI-Movie Experiment
|
|
Author(s):
|
Marco Antonio Pinto-Orellana* and Chee-Ming Ting and Jeremy Skipper and Steven Small and Hernando Ombao
|
|
Companies:
|
Statistics, CEMSE Division. King Abdullah University of Science and Technology and King Abdullah University of Science and Technology and Institute for Multimodal Communication. University College London and University of California, Irvine and King Abdullah University of Science and Technology
|
|
Keywords:
|
Brain connectivity;
Switching-regime vector autoregressive model;
Brain states;
Software toolbox;
Multiple stimuli
|
|
Abstract:
|
The goal of this talk is to present a model for changes in brain connectivity while a participant is watching a short movie. Our proposed model is a switching-regime vector autoregressive model (S-VAR) that is fitted over clustered time-varying vector autoregressive (TV-VAR) coefficients. This model captures dynamic changes in brain states (or brain connectivity) as the participant is exposed to different types of audio-visual and cognitive stimuli while watching a movie. The experiment setting promotes a continuous interaction between the stimuli, and consequently, the states. In our model, those transitions between states are modeled by probabilities that are allowed to depend on the combinations of previously present stimuli. As part of this project, one contribution is a computer toolbox which implements seamless integration between model estimation and visualization of results.
|
Authors who are presenting talks have a * after their name.
