Abstract | ||
---|---|---|
Functional Magnetic Resonance Imaging (fMRI) gives us a unique insight into the processes of the brain, and opens up for analyzing the functional activation patterns of the underlying sources. Task-inferred supervised learning with restrictive assumptions in the regression set-up, restricts the exploratory nature of the analysis. Fully unsupervised independent component analysis (ICA) algorithms, on the other hand, can struggle to detect clear classifiable components on single-subject data. We attribute this shortcoming to inadequate modeling of the fMRI source signals by failing to incorporate its temporal nature. fMRI source signals, biological stimuli and non-stimuli-related artifacts are all smooth over a time-scale compatible with the sampling time (TR). We therefore propose Gaussian process ICA (GPICA), which facilitates temporal dependency by the use of Gaussian process source priors. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1016/j.neuroimage.2017.02.070 | NeuroImage |
Keywords | Field | DocType |
Gaussian processes,FMRI,Source separation,Independent component analysis,Convolutive mixing,Bayesian inference | Bayesian inference,Pattern recognition,Markov chain Monte Carlo,Computer science,Supervised learning,Gaussian process,Independent component analysis,Artificial intelligence,Hierarchical database model,Source separation,Kernel (statistics) | Journal |
Volume | ISSN | Citations |
152 | 1053-8119 | 0 |
PageRank | References | Authors |
0.34 | 16 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ditte Høvenhoff Hald | 1 | 0 | 0.34 |
Ricardo Henao | 2 | 286 | 23.85 |
Winther, Ole | 3 | 960 | 106.57 |