A probabilistic model for discovering high level brain activities from fMRI

Publication Type:
Conference Proceeding
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2011, 7062 LNCS (PART 1), pp. 329 - 336
Issue Date:
Full metadata record
Files in This Item:
Filename Description Size
Thumbnail2011001467OK_Li.pdf389.59 kB
Adobe PDF
Functional magnetic resonance imaging (fMRI) has provided an invaluable method of investing real time neuron activities. Statistical tools have been developed to recognise the mental state from a batch of fMRI observations over a period. However, an interesting question is whether it is possible to estimate the real time mental states at each moment during the fMRI observation. In this paper, we address this problem by building a probabilistic model of the brain activity. We model the tempo-spatial relations among the hidden high-level mental states and observable low-level neuron activities. We verify our model by experiments on practical fMRI data. The model also implies interesting clues on the task-responsible regions in the brain. © 2011 Springer-Verlag.
Please use this identifier to cite or link to this item: