Statistical analysis of brain tissue images in the wavelet domain: wavelet-based morphometry

Toward individualized connectomes of brain morphology

The morphological brain connectome (MBC) delineates the coordinated patterns of local morphological features (such as cortical thickness) across brain regions. While classically constructed using population-based approaches, there is a growing trend toward individualized modeling. Currently, the methods for individualized MBCs are varied, posing challenges for method selection and cross-study comparisons. Here, we summarize how individualized MBCs are modeled through low-order methods (correlation-, divergence-, distance-, and deviation-based methods) describing relations in brain morphology, as well as high-order methods capturing similarities in these low-order relations. We discuss the merits and limitations of different methods, examining them in the context of robustness, reproducibility, and reliability. We highlight the importance of elucidating the cellular and molecular mechanisms underlying the individualized connectome, and establishing normative benchmarks to assess individual variation in development, aging, and neuropsychiatric disorders.

Mapping white matter structural covariance connectivity for single subject using wavelet transform with T1-weighted anatomical brain MRI

Introduction

Current studies of structural covariance networks were focused on the gray matter in the human brain. The structural covariance connectivity in the white matter remains largely unexplored. This paper aimed to build novel metrics that can infer white matter structural covariance connectivity, and to explore the predictive power of the proposed features.

Methods

To this end, a cohort of 315 adult subjects with the anatomical brain MRI datasets were obtained from the publicly available Dallas Lifespan Brain Study (DLBS) project. The 3D wavelet transform was applied on the individual voxel-based morphology (VBM) volume to obtain the white matter structural covariance connectivity. The predictive models for cognitive functions were built using support vector regression (SVR).

Results

The predictive models exhibited comparable performance with previous studies. The novel features successfully predicted the individual ability of digit comparison (DC) (r = 0.41 ± 0.01, p < 0.01) and digit symbol (DSYM) (r = 0.5 ± 0.01, p < 0.01). The sensorimotor-related white matter system exhibited as the most predictive network node. Furthermore, the node strengths of sensorimotor mode were significantly correlated to cognitive scores.

Discussion

The results suggested that the white matter structural covariance connectivity was informative and had potential for predictive tasks of brain-behavior research.

Mapping individual voxel-wise morphological connectivity using wavelet transform of voxel-based morphology

Mapping individual brain networks has drawn significant research interest in recent years. Most individual brain networks developed to date have been based on fMRI or diffusion MRI. Given recent concerns regarding confounding artifacts, various preprocessing steps are generally included in functional or structural brain networks. Notably, voxel-based morphometry (VBM) derived from anatomical MRI exhibits high signal-to-noise ratios and has been applied to individual interregional morphological networks. To the best of our knowledge, individual voxel-wise morphological networks remain unexplored. The goal of this research is twofold: to build novel metrics for individual voxel-wise morphological networks and to test the reliability of the proposed morphological connectivity. To this end, anatomical scans of a cohort of healthy subjects were obtained from a public database. The anatomical datasets were preprocessed and normalized to the standard brain space. For each individual, wavelet-transform was applied on the VBM measures to obtain voxel-wise hierarchical features. The voxel-wise morphological connectivity was computed based on the wavelet features. Reliable brain hubs were detected by the z-scores of degree centrality. High reliability was discovered by test-retest analysis. No effects of wavelet scale, network threshold or network type were found on hubs of group-level degree centrality. However, significant effects of wavelet scale, network threshold and network type were found on individual-level degree centrality. Significant effects of network threshold and network type were found on reliability of degree centrality. The results suggested that the voxel-wise morphological connectivity was reliable and exhibited a hub structure. Moreover, the voxel-wise morphological connectivity could reflect individual differences. In summary, individual voxel-wise wavelet-based features can probe morphological connectivity and may be beneficial for investigating the brain morphological connectomes.

Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis

A relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discrimination between both psychotic diagnostics. To assess the efficacy of sMRI data for diagnostic prediction in psychosis we objectively evaluated the discriminative power of a wide range of commonly used machine learning algorithms (ridge, lasso, elastic net and L0 norm regularized logistic regressions, a support vector classifier, regularized discriminant analysis, random forests and a Gaussian process classifier) on main sMRI features including grey and white matter voxel-based morphometry (VBM), vertex-based cortical thickness and volume, region of interest volumetric measures and wavelet-based morphometry (WBM) maps. All possible combinations of algorithms and data features were considered in pairwise classifications of matched samples of healthy controls (N = 127), patients with schizophrenia (N = 128) and patients with bipolar disorder (N = 128). Results show that the selection of feature type is important, with grey matter VBM (without data reduction) delivering the best diagnostic prediction rates (averaging over classifiers: schizophrenia vs. healthy 75%, bipolar disorder vs. healthy 63% and schizophrenia vs. bipolar disorder 62%) whereas algorithms usually yielded very similar results. Indeed, those grey matter VBM accuracy rates were not even improved by combining all feature types in a single prediction model. Further multi-class classifications considering the three groups simultaneously made evident a lack of predictive power for the bipolar group, probably due to its intermediate anatomical features, located between those observed in healthy controls and those found in patients with schizophrenia. Finally, we provide MRIPredict (https://www.nitrc.org/projects/mripredict/), a free tool for SPM, FSL and R, to easily carry out voxelwise predictions based on VBM images.

Anatomic and Functional Connectivity Relationship in Autistic Children During Three Different Experimental Conditions

A group of 21 autistic children were studied for determining the relationship between the anatomic (AC) versus functional (FC) connectivity, considering short-range and long-range brain networks. AC was assessed by the DW-MRI technique and FC by EEG coherence calculation, in three experimental conditions: basal, watching a popular cartoon with audio (V-A), and with muted audio track (VwA). For short-range connections, basal records, statistical significant correlations were found for all EEG bands in the left hemisphere, but no significant correlations were found for fast EEG frequencies in the right hemisphere. For the V-A condition, significant correlations were mainly diminished for the left hemisphere; for the right hemisphere, no significant correlations were found for the fast EEG frequency bands. For the VwA condition, significant correlations for the rapid EEG frequencies mainly disappeared for the right hemisphere. For long-range connections, basal records showed similar correlations for both hemispheres. For the right hemisphere, significant correlations incremented to all EEG bands for the V-A condition, but these significant correlations disappeared for the fast EEG frequencies in the VwA condition. It appears that in a resting-state condition, AC is better associated with functional connectivity for short-range connections in the left hemisphere. The V-A experimental condition enriches the AC and FC association for long-range connections in the right hemisphere. This might be related to an effective connectivity improvement due to full video stimulation (visual and auditory). An impaired audiovisual interaction in the right hemisphere might explain why significant correlations disappeared for the fast EEG frequencies in the VwA experimental condition.

Anatomically-adapted graph wavelets for improved group-level fMRI activation mapping

A graph based framework for fMRI brain activation mapping is presented. The approach exploits the spectral graph wavelet transform (SGWT) for the purpose of defining an advanced multi-resolutional spatial transformation for fMRI data. The framework extends wavelet based SPM (WSPM), which is an alternative to the conventional approach of statistical parametric mapping (SPM), and is developed specifically for group-level analysis. We present a novel procedure for constructing brain graphs, with subgraphs that separately encode the structural connectivity of the cerebral and cerebellar gray matter (GM), and address the inter-subject GM variability by the use of template GM representations. Graph wavelets tailored to the convoluted boundaries of GM are then constructed as a means to implement a GM-based spatial transformation on fMRI data. The proposed approach is evaluated using real as well as semi-synthetic multi-subject data. Compared to SPM and WSPM using classical wavelets, the proposed approach shows superior type-I error control. The results on real data suggest a higher detection sensitivity as well as the capability to capture subtle, connected patterns of brain activity.

Multi-resolution statistical analysis of brain connectivity graphs in preclinical Alzheimer's disease

There is significant interest, both from basic and applied research perspectives, in understanding how structural/functional connectivity changes can explain behavioral symptoms and predict decline in neurodegenerative diseases such as Alzheimer's disease (AD). The first step in most such analyses is to encode the connectivity information as a graph; then, one may perform statistical inference on various 'global' graph theoretic summary measures (e.g., modularity, graph diameter) and/or at the level of individual edges (or connections). For AD in particular, clear differences in connectivity at the dementia stage of the disease (relative to healthy controls) have been identified. Despite such findings, AD-related connectivity changes in preclinical disease remain poorly characterized. Such preclinical datasets are typically smaller and group differences are weaker. In this paper, we propose a new multi-resolution method for performing statistical analysis of connectivity networks/graphs derived from neuroimaging data. At the high level, the method occupies the middle ground between the two contrasts - that is, to analyze global graph summary measures (global) or connectivity strengths or correlations for individual edges similar to voxel based analysis (local). Instead, our strategy derives a Wavelet representation at each primitive (connection edge) which captures the graph context at multiple resolutions. We provide extensive empirical evidence of how this framework offers improved statistical power by analyzing two distinct AD datasets. Here, connectivity is derived from diffusion tensor magnetic resonance images by running a tractography routine. We first present results showing significant connectivity differences between AD patients and controls that were not evident using standard approaches. Later, we show results on populations that are not diagnosed with AD but have a positive family history risk of AD where our algorithm helps in identifying potentially subtle differences between patient groups. We also give an easy to deploy open source implementation of the algorithm for use within studies of connectivity in AD and other neurodegenerative disorders.

Networks of anatomical covariance

Functional imaging or diffusion-weighted imaging techniques are widely used to understand brain connectivity at the systems level and its relation to normal neurodevelopment, cognition or brain disorders. It is also possible to extract information about brain connectivity from the covariance of morphological metrics derived from anatomical MRI. These covariance patterns may arise from genetic influences on normal development and aging, from mutual trophic reinforcement as well as from experience-related plasticity. This review describes the basic methodological strategies, the biological basis of the observed covariance as well as applications in normal brain and brain disease before a final review of future prospects for the technique.