Volume 13 Supplement 1

Twenty First Annual Computational Neuroscience Meeting: CNS*2012

Open Access

Dynamic Bayesian network modeling for intervention mechanism

BMC Neuroscience201213(Suppl 1):P24


Published: 16 July 2012

Functional magnetic resonance imaging (fMRI) is an important experimental tool in neuroscience. However, how to analyze the fMRI data effectively and accurately has become one of the major challenges in computational neuroscience [1]. Integrative body-mind training (IBMT) was adopted from traditional Chinese medicine and has been proven to improve attention and self-regulation compared to same amount of relaxation training using regular neuroimaging analysis methods in our previous studies [2, 3]. The greatest advantage of dynamic Bayesian networks (DBNs) is that it could demonstrate the temporal and causal relationships among different brain regions more accurately. We here propose DBNs to identify the brain changes using the fMRI data sets of five days of IBMT intervention. At first, we employed Statistical Parametric Mapping software (SPM8, http://www.fil.ion.ucl.ac.uk/spm) to preprocess the images. Second, the Markov chain was introduced to model the fMRI time-series and obtained the temporal relationships among brain regions. Third, we used K2 algorithm to learn the structure of the DBNs and adopted the greedy search algorithm to search for the local best optimal connectivity structure from fMRI data [4]. Finally, we obtained the DBNs of the IBMT group and relaxation group, which represent the interactions among brain regions with temporal processes. The nodes in the DBN represented the activations of brain regions at a specific time while the edges denoted the connectivity strengths between brain regions. The DBNs of IBMT group was different from that of relaxation group in the several brain regions particularly in the anterior cingulated cortex (ACC), which was consistent with our previous research findings [3]. The DBNs is an efficient method to demonstrate the brain mechanism of short-term meditation intervention.



This work was supported by the NSFC 60971096, Office of Naval Research and R21DA030066.

Authors’ Affiliations

Research Center of Psychological Development and Education, Liaoning Normal University
Psychology Department of Education School, Liaoning Normal University
Texas Tech Neuroimaging Institute and Department of Psychology, Texas Tech University
Institute of Neuroinformatics, Dalian University of Technology


  1. Chen R, Resnick SM, Davatzikos C, Herskovits EH: Dynamic Bayesian network modeling for longitudinal brain morphometry. NeuroImage. 2012, 59: 2330-2338. 10.1016/j.neuroimage.2011.09.023.PubMed CentralView ArticlePubMedGoogle Scholar
  2. Tang YY, Ma YH, Wang JH, et al: Short term meditation training improves attention and self-regulation. Proceedings of the National Academy of Sciences. 2007, 104: 17152-17156. 10.1073/pnas.0707678104.View ArticleGoogle Scholar
  3. Tang YY, Ma YH, Fan YX, et al: Central and autonomic nervous system interaction is altered by short-term meditation. Proceedings of the National Academy of Sciences. 2009, 106: 8865-8870. 10.1073/pnas.0904031106.View ArticleGoogle Scholar
  4. Cooper GF, Herskovits EH: A Bayesian method for the induction of probabilistic networks from data. Machine Learning. 1992, 9: 309-347.Google Scholar


© Sun and Tang; licensee BioMed Central Ltd. 2012

This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.