Discrete dynamic Bayesian network analysis of fMRI data

• Published in: Human brain mapping Vol. 30; no. 1; pp. 122 - 137
• Main Authors: Burge, John, Lane, Terran, Link, Hamilton, Qiu, Shibin, Clark, Vincent P.
• Format: Journal Article
• Language: English
• Published: Hoboken Wiley Subscription Services, Inc., A Wiley Company 01.01.2009; Wiley-Liss
• Subjects: Aged; Algorithms; Alzheimer Disease - pathology; Alzheimer Disease - physiopathology; amygdala; Bayes Theorem; Bayesian networks; Biological and medical sciences; Brain - anatomy & histology; Brain - physiology; Brain Mapping - methods; Cerebrovascular Circulation - physiology; Data Interpretation, Statistical; dementia; Electrodiagnosis. Electric activity recording; Fourier Analysis; functional connectivity; Humans; Image Processing, Computer-Assisted - methods; Investigative techniques, diagnostic techniques (general aspects); Magnetic Resonance Imaging - methods; Medical sciences; Nerve Net - anatomy & histology; Nerve Net - physiology; Nervous system; Neural Networks, Computer; nonlinear analysis; Normal Distribution; Radiodiagnosis. Nmr imagery. Nmr spectrometry; Signal Processing, Computer-Assisted; Talairach atlas; Atlas; Data analysis; Nervous system diseases; nonlinear analysis; Amygdala; Radiodiagnosis; Nuclear magnetic resonance imaging; functional connectivity; Talairach atlas; Nonlinearity; Degenerative disease; Bayesian networks; Dementia
• ISSN: 1065-9471; 1097-0193; 1097-0193
• DOI: 10.1002/hbm.20490

We examine the efficacy of using discrete Dynamic Bayesian Networks (dDBNs), a data‐driven modeling technique employed in machine learning, to identify functional correlations among neuroanatomical regions of interest. Unlike many neuroimaging analysis techniques, this method is not limited by linear and/or Gaussian noise assumptions. It achieves this by modeling the time series of neuroanatomical regions as discrete, as opposed to continuous, random variables with multinomial distributions. We demonstrated this method using an fMRI dataset collected from healthy and demented elderly subjects (Buckner, et al., 2000: J Cogn Neurosci 12:24‐34) and identify correlates based on a diagnosis of dementia. The results are validated in three ways. First, the elicited correlates are shown to be robust over leave‐one‐out cross‐validation and, via a Fourier bootstrapping method, that they were not likely due to random chance. Second, the dDBNs identified correlates that would be expected given the experimental paradigm. Third, the dDBN's ability to predict dementia is competitive with two commonly employed machine‐learning classifiers: the support vector machine and the Gaussian naïve Bayesian network. We also verify that the dDBN selects correlates based on non‐linear criteria. Finally, we provide a brief analysis of the correlates elicited from Buckner et al.'s data that suggests that demented elderly subjects have reduced involvement of entorhinal and occipital cortex and greater involvement of the parietal lobe and amygdala in brain activity compared with healthy elderly (as measured via functional correlations among BOLD measurements). Limitations and extensions to the dDBN method are discussed. Hum Brain Mapp, 2009. © 2007 Wiley‐Liss, Inc.