Conference paper
Identification of non-linear models of neural activity in bold fmri
Non-linear hemodynamic models express the BOLD signal as a nonlinear, parametric functional of the temporal sequence of local neural activity. Several models have been proposed for this neural activity. We identify one such parametric model by estimating the distribution of its parameters. These distributions are themselves stochastic, therefore we estimate their variance by epoch based leave-one-out cross validation, using a Metropolis-Hastings algorithm for sampling of the posterior parameter distribution.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2006 |
Pages: | 952-955 |
Proceedings: | 2006 IEEE International Symposium on Biomedical Imaging |
ISBN: | 078039576X , 078039576x and 9780780395763 |
Types: | Conference paper |
DOI: | 10.1109/ISBI.2006.1625077 |
ORCIDs: | Madsen, Kristoffer Hougaard and Hansen, Lars Kai |
Blood Differential equations Hemodynamics Jacobian matrices Metropolis-Hastings algorithm Neuroimaging Noise measurement Parametric statistics Sampling methods Signal processing Stochastic processes biomedical MRI epoch based leave-one-out cross validation haemodynamics medical image processing neurophysiology nonlinear hemodynamic models parameter estimation posterior parameter distribution stochastic distributions stochastic processes temporal neural activity sequence variance estimation