Sparse Representation of Brain Aging: Extracting Covariance Patterns from Structural MRI

Abstract

An enhanced understanding of how normal aging alters brain structure is urgently needed for the early diagnosis and treatment of age-related mental diseases. Structural magnetic resonance imaging (MRI) is a reliable technique used to detect age-related changes in the human brain. Currently, multivariate pattern analysis (MVPA) enables the exploration of subtle and distributed changes of data obtained from structural MRI images. In this study, a new MVPA approach based on sparse representation has been employed to investigate the anatomical covariance patterns of normal aging. Two groups of participants (group 1∶290 participants; group 2∶56 participants) were evaluated in this study. These two groups were scanned with two 1.5 T MRI machines. In the first group, we obtained the discriminative patterns using a t-test filter and sparse representation step. We were able to distinguish the young from old cohort with a very high accuracy using only a few voxels of the discriminative patterns (group 1∶98.4%; group 2∶96.4%). The experimental results showed that the selected voxels may be categorized into two components according to the two steps in the proposed method. The first component focuses on the precentral and postcentral gyri, and the caudate nucleus, which play an important role in sensorimotor tasks. The strongest volume reduction with age was observed in these clusters. The second component is mainly distributed over the cerebellum, thalamus, and right inferior frontal gyrus. These regions are not only critical nodes of the sensorimotor circuitry but also the cognitive circuitry although their volume shows a relative resilience against aging. Considering the voxels selection procedure, we suggest that the aging of the sensorimotor and cognitive brain regions identified in this study has a covarying relationship with each other.

Influence of Resting-State Network on Lateralization of Functional Connectivity in Mesial Temporal Lobe Epilepsy

“…28 The support vector machine recursive feature elimination method can predict group membership at an individual subject level, and the results obtained by using this method may be clinically useful 29 because the results can include unique information that may be overlooked by univariate voxel-based morphometry approaches. 30,31 As indicated in Fig 1, maximal classification accuracy (generalization rate ϭ 95.2%, area under the receiver operating characteristic curve ϭ 100%) was obtained by using only 23 features. We assigned connection strengths on the basis of the occurrence of these connections in the leave-one-out cross-validation results (Fig 3).…”

Imaging structural co-variance between human brain regions

“…Classification studies have attempted to find MRI signatures that are associated with particular diseases or outcomes and that could be useful in diagnosis or treatment planning. Here, the inter-dependence of the anatomy of brain regions is both a hurdle (because many classification techniques are optimized for datasets of independent features) and an opportunity (because deviations from expected co-variance patterns provide additional information), and successful preliminary studies have explicitly or implicitly incorporated brain co-variance structure into their predictive models 184–186 . The complex mixture of developmental, genetic and environmental factors that influence anatomical structure and inter-regional dependence even in healthy individuals may be one reason why the clinical utility of MRI in psychiatry has yet to meet expectations.…”

Age estimation using cortical surface pattern combining thickness with curvatures

“…Dosenbach et al [6] performed a binary SVM classification of individual as either children or adults to assess the relative functional brain maturity and reached the accuracy of 91%. A new MVPA approach based on sparse representation has been employed to investigate the anatomical covariance patterns of normal aging by classifying two age groups in [28] and the mean accuracy is about 97.4%. Comparing with previous studies, the age groups classification with our CSP yields a high CV accuracy (97.77%) and a high sensitivity/specificity (97.30%/98.10%).…”