TY - GEN
T1 - Determining effective connectivity from fMRI data using a Gaussian dynamic Bayesian network
AU - Wu, Xia
AU - Li, Juan
AU - Yao, Li
PY - 2012
Y1 - 2012
N2 - Two techniques that are based on the Bayesian network, Gaussian Bayesian network (BN) and discrete dynamic Bayesian network (DBN), have recently been used to determine the effective connectivity from functional magnetic resonance imaging (fMRI) data in an exploratory manner and provide a new method for the interactions among brain regions. However, Gaussian BN ignores the temporal relationships of interactions among brain regions, while discrete DBN loses a great of information by discretizing data. In this study, we proposed Gaussian DBN, which is based on Gaussian assumptions, to capture the temporal characteristics of connectivity with less associated loss of information. Synthetic data were generated to validate the effectiveness of this method, and the results were compared with discrete DBN. The result demonstrated that our method is both more robust than discrete DBN and an improvement over BN.
AB - Two techniques that are based on the Bayesian network, Gaussian Bayesian network (BN) and discrete dynamic Bayesian network (DBN), have recently been used to determine the effective connectivity from functional magnetic resonance imaging (fMRI) data in an exploratory manner and provide a new method for the interactions among brain regions. However, Gaussian BN ignores the temporal relationships of interactions among brain regions, while discrete DBN loses a great of information by discretizing data. In this study, we proposed Gaussian DBN, which is based on Gaussian assumptions, to capture the temporal characteristics of connectivity with less associated loss of information. Synthetic data were generated to validate the effectiveness of this method, and the results were compared with discrete DBN. The result demonstrated that our method is both more robust than discrete DBN and an improvement over BN.
KW - Dynamic Bayesian network (DBN)
KW - Effective connectivity
KW - FMRI
UR - http://www.scopus.com/inward/record.url?scp=84869010944&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-34475-6_5
DO - 10.1007/978-3-642-34475-6_5
M3 - Conference contribution
AN - SCOPUS:84869010944
SN - 9783642344749
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 33
EP - 39
BT - Neural Information Processing - 19th International Conference, ICONIP 2012, Proceedings
T2 - 19th International Conference on Neural Information Processing, ICONIP 2012
Y2 - 12 November 2012 through 15 November 2012
ER -