Learning brain connectivity of Alzheimer's disease by sparse inverse covariance estimation

Abstract

Rapid advances in neuroimaging techniques provide great potentials for study of Alzheimer’s disease (AD). Existing findings have shown that AD is closely related to alteration in the functional brain network, i.e., the functional connectivity between different brain regions. In this paper, we propose a method based on sparse inverse covariance estimation (SICE) to identify functional brain connectivity networks from PET data. Our method is able to identify both the connectivity network structure and strength for a large number o f brain regions with small sample sizes. We apply the proposed method to the PET data of AD, mild cognitive impairment (MCI), and normal control (NC) subjects. Compared with NC, AD shows decrease in the amount of inter-region functional connectivity within the temporal lobe especially between the area around hippocampus and other regions and increase in the amount of connectivity within the frontal lobe as well as between the parietal and occipital lobes. Also, AD shows weaker between-lobe connectivity than within-lobe connectivity and weaker between-hemisphere connectivity, compared with NC. In addition to being a method for knowledge discovery about AD, the proposed SICE method can also be used for classifying new subjects, which makes it a suitable approach for novel connectivity-based AD biomarker identification. Our experiments show that the best sensitivity and specificity our method can achieve in AD vs. NC classification are 88% and 88%, respectively.

Metabolic and structural connectivity within the default mode network relates to working memory performance in young healthy adults

“…First, metabolic connectivity as presently quantified allows reasonable inferences only at a group level. However, recent advances in functional connectivity modeling have allowed identifying brain connectivity networks from PET data at a single subject level (Huang et al, 2010). Once a group-level based connectivity model is available, the method enables to classify new subjects on the basis of their individual connectivity profile.…”

“…Thus, investigation of connectivity with (FDG-) PET has been considered as principally possible at a group level only. Yet, very recent evidence indicates encouraging properties of FDG-PET-based resting state connectivity at an individual level (Huang et al, 2010;Toussaint et al, 2012). To differentiate from functional connectivity as quantified with fMRI, here we use the term "metabolic connectivity" suggested by Lee et al (2008).…”

Bayesian inference of structural brain networks

“…Hence, finding its inverse requires suboptimal solutions such as the generalized inverse or pseudo-inverse (Ryali et al, 2011). As a solution to this problem, regularization approaches have been suggested to find sparse approximations of the inverse covariance matrix (Friedman et al, 2008;Huang et al, 2010). In our setup, the sparsity structure is readily available in the form of structural connectivity A.…”

Inferring functional connectivity in fMRI using minimum partial correlation

“…Several other approaches were also proposed for overcoming this drawback. Utilizing the monotone property of the lasso path of ICOV, Huang et al [20] used the largest regularization parameter value that preserves the existence of a connection as a quasi-measure for the strength of this connection. Recently, a statistic also based on the lasso path was proposed to test whether a regularised inverse covariance matrix represents all the real connections [21,22] .…”