Jing Sui

A review of multivariate methods for multimodal fusion of brain imaging data

The development of various neuroimaging techniques is rapidly improving the measurements of brain function/structure. However, despite improvements in individual modalities, it is becoming increasingly clear that the most effective research approaches will utilize multi-modal fusion, which takes advantage of the fact that each modality provides a limited view of the brain. The goal of multi-modal fusion is to capitalize on the strength of each modality in a joint analysis, rather than a separate analysis of each. This is a more complicated endeavor that must be approached more carefully and efficient methods should be developed to draw generalized and valid conclusions from high dimensional data with a limited number of subjects. Numerous research efforts have been reported in the field based on various statistical approaches, e.g. independent component analysis (ICA), canonical correlation analysis (CCA) and partial least squares (PLS). In this review paper, we survey a number of multivariate methods appearing in previous multimodal fusion reports, mostly fMRI with other modality, which were performed with or without prior information. A table for comparing optimization assumptions, purpose of the analysis, the need of priors, dimension reduction strategies and input data types is provided, which may serve as a valuable reference that helps readers understand the trade-offs of the 7 methods comprehensively. Finally, we evaluate 3 representative methods via simulation and give some suggestions on how to select an appropriate method based on a given research.

Single subject prediction of brain disorders in neuroimaging: Promises and pitfalls

ABSTRACT Neuroimaging‐based single subject prediction of brain disorders has gained increasing attention in recent years. Using a variety of neuroimaging modalities such as structural, functional and diffusion MRI, along with machine learning techniques, hundreds of studies have been carried out for accurate classification of patients with heterogeneous mental and neurodegenerative disorders such as schizophrenia and Alzheimers disease. More than 500 studies have been published during the past quarter century on single subject prediction focused on a multiple brain disorders. In the first part of this study, we provide a survey of more than 200 reports in this field with a focus on schizophrenia, mild cognitive impairment (MCI), Alzheimers disease (AD), depressive disorders, autism spectrum disease (ASD) and attention‐deficit hyperactivity disorder (ADHD). Detailed information about those studies such as sample size, type and number of extracted features and reported accuracy are summarized and discussed. To our knowledge, this is by far the most comprehensive review of neuroimaging‐based single subject prediction of brain disorders. In the second part, we present our opinion on major pitfalls of those studies from a machine learning point of view. Common biases are discussed and suggestions are provided. Moreover, emerging trends such as decentralized data sharing, multimodal brain imaging, differential diagnosis, disease subtype classification and deep learning are also discussed. Based on this survey, there is extensive evidence showing the great potential of neuroimaging data for single subject prediction of various disorders. However, the main bottleneck of this exciting field is still the limited sample size, which could be potentially addressed by modern data sharing models such as the ones discussed in this paper. Emerging big data technologies and advanced data‐intensive machine learning methodologies such as deep learning have coincided with an increasing need for accurate, robust and generalizable single subject prediction of brain disorders during an exciting time. In this report, we survey the past and offer some opinions regarding the road ahead. HighlightsPast efforts on classification of brain disorders are comprehensively reviewed.The common pitfalls from machine learning point of view are discussed.Emerging trends related to single‐subject prediction are reviewed and discussed.

Exploring the psychosis functional connectome: aberrant intrinsic networks in schizophrenia and bipolar disorder

Intrinsic functional brain networks (INs) are regions showing temporal coherence with one another. These INs are present in the context of a task (as opposed to an undirected task such as rest), albeit modulated to a degree both spatially and temporally. Prominent networks include the default mode, attentional fronto-parietal, executive control, bilateral temporal lobe, and motor networks. The characterization of INs has recently gained considerable momentum, however; most previous studies evaluate only a small subset of the INs (e.g., default mode). In this paper we use independent component analysis to study INs decomposed from functional magnetic resonance imaging data collected in a large group of schizophrenia patients, healthy controls, and individuals with bipolar disorder, while performing an auditory oddball task. Schizophrenia and bipolar disorder share significant overlap in clinical symptoms, brain characteristics, and risk genes which motivates our goal of identifying whether functional imaging data can differentiate the two disorders. We tested for group differences in properties of all identified INs including spatial maps, spectra, and functional network connectivity. A small set of default mode, temporal lobe, and frontal networks with default mode regions appearing to play a key role in all comparisons. Bipolar subjects showed more prominent changes in ventromedial and prefrontal default mode regions whereas schizophrenia patients showed changes in posterior default mode regions. Anti-correlations between left parietal areas and dorsolateral prefrontal cortical areas were different in bipolar and schizophrenia patients and amplitude was significantly different from healthy controls in both patient groups. Patients exhibited similar frequency behavior across multiple networks with decreased low frequency power. In summary, a comprehensive analysis of INs reveals a key role for the default mode in both schizophrenia and bipolar disorder.

Altered Topological Properties of Functional Network Connectivity in Schizophrenia during Resting State: A Small-World Brain Network Study

Aberrant topological properties of small-world human brain networks in patients with schizophrenia (SZ) have been documented in previous neuroimaging studies. Aberrant functional network connectivity (FNC, temporal relationships among independent component time courses) has also been found in SZ by a previous resting state functional magnetic resonance imaging (fMRI) study. However, no study has yet determined if topological properties of FNC are also altered in SZ. In this study, small-world network metrics of FNC during the resting state were examined in both healthy controls (HCs) and SZ subjects. FMRI data were obtained from 19 HCs and 19 SZ. Brain images were decomposed into independent components (ICs) by group independent component analysis (ICA). FNC maps were constructed via a partial correlation analysis of ICA time courses. A set of undirected graphs were built by thresholding the FNC maps and the small-world network metrics of these maps were evaluated. Our results demonstrated significantly altered topological properties of FNC in SZ relative to controls. In addition, topological measures of many ICs involving frontal, parietal, occipital and cerebellar areas were altered in SZ relative to controls. Specifically, topological measures of whole network and specific components in SZ were correlated with scores on the negative symptom scale of the Positive and Negative Symptom Scale (PANSS). These findings suggest that aberrant architecture of small-world brain topology in SZ consists of ICA temporally coherent brain networks.

An ICA-based method for the identification of optimal FMRI features and components using combined group-discriminative techniques

Extraction of relevant features from multitask functional MRI (fMRI) data in order to identify potential biomarkers for disease, is an attractive goal. In this paper, we introduce a novel feature-based framework, which is sensitive and accurate in detecting group differences (e.g. controls vs. patients) by proposing three key ideas. First, we integrate two goal-directed techniques: coefficient-constrained independent component analysis (CC-ICA) and principal component analysis with reference (PCA-R), both of which improve sensitivity to group differences. Secondly, an automated artifact-removal method is developed for selecting components of interest derived from CC-ICA, with an average accuracy of 91%. Finally, we propose a strategy for optimal feature/component selection, aiming to identify optimal group-discriminative brain networks as well as the tasks within which these circuits are engaged. The group-discriminating performance is evaluated on 15 fMRI feature combinations (5 single features and 10 joint features) collected from 28 healthy control subjects and 25 schizophrenia patients. Results show that a feature from a sensorimotor task and a joint feature from a Sternberg working memory (probe) task and an auditory oddball (target) task are the top two feature combinations distinguishing groups. We identified three optimal features that best separate patients from controls, including brain networks consisting of temporal lobe, default mode and occipital lobe circuits, which when grouped together provide improved capability in classifying group membership. The proposed framework provides a general approach for selecting optimal brain networks which may serve as potential biomarkers of several brain diseases and thus has wide applicability in the neuroimaging research community.

Assessing dynamic brain graphs of time-varying connectivity in fMRI data: application to healthy controls and patients with schizophrenia

Graph theory-based analysis has been widely employed in brain imaging studies, and altered topological properties of brain connectivity have emerged as important features of mental diseases such as schizophrenia. However, most previous studies have focused on graph metrics of stationary brain graphs, ignoring that brain connectivity exhibits fluctuations over time. Here we develop a new framework for accessing dynamic graph properties of time-varying functional brain connectivity in resting-state fMRI data and apply it to healthy controls (HCs) and patients with schizophrenia (SZs). Specifically, nodes of brain graphs are defined by intrinsic connectivity networks (ICNs) identified by group independent component analysis (ICA). Dynamic graph metrics of the time-varying brain connectivity estimated by the correlation of sliding time-windowed ICA time courses of ICNs are calculated. First- and second-level connectivity states are detected based on the correlation of nodal connectivity strength between time-varying brain graphs. Our results indicate that SZs show decreased variance in the dynamic graph metrics. Consistent with prior stationary functional brain connectivity works, graph measures of identified first-level connectivity states show lower values in SZs. In addition, more first-level connectivity states are disassociated with the second-level connectivity state which resembles the stationary connectivity pattern computed by the entire scan. Collectively, the findings provide new evidence about altered dynamic brain graphs in schizophrenia, which may underscore the abnormal brain performance in this mental illness.

Brain connectivity networks in schizophrenia underlying resting state functional magnetic resonance imaging

Schizophrenia (SZ) is a severe neuropsychiatric disorder. A leading hypothesis is that SZ is a brain dysconnection syndrome, involving abnormal interactions between widespread brain networks. Resting state functional magnetic resonance imaging (R-fMRI) is a powerful tool to explore the dysconnectivity of brain networks in SZ and other disorders. Seed-based functional connectivity analysis, spatial independent component analysis (ICA), and graph theory-based analysis are popular methods to quantify brain network connectivity in R-fMRI data. Widespread network dysconnectivity in SZ has been observed using both seed-based analysis and ICA, although most seed-based studies report decreased connectivity while ICA studies report both increases and decreases. Importantly, most of the findings from both techniques are also associated with typical symptoms of the illness. Disrupted topological properties and altered modular community structure of brain system in SZ have been shown using graph theory-based analysis. Overall, the resting-state findings regarding brain networks deficits have advanced our understanding of the underlying pathology of SZ. In this article, we review aberrant brain connectivity networks in SZ measured in R-fMRI by the above approaches, and discuss future challenges.

Function-structure associations of the brain: Evidence from multimodal connectivity and covariance studies

Despite significant advances in multimodal imaging techniques and analysis approaches, unimodal studies are still the predominant way to investigate brain changes or group differences, including structural magnetic resonance imaging (sMRI), functional MRI (fMRI), diffusion tensor imaging (DTI) and electroencephalography (EEG). Multimodal brain studies can be used to understand the complex interplay of anatomical, functional and physiological brain alterations or development, and to better comprehend the biological significance of multiple imaging measures. To examine the function-structure associations of the brain in a more comprehensive and integrated manner, we reviewed a number of multimodal studies that combined two or more functional (fMRI and/or EEG) and structural (sMRI and/or DTI) modalities. In this review paper, we specifically focused on multimodal neuroimaging studies on cognition, aging, disease and behavior. We also compared multiple analysis approaches, including univariate and multivariate methods. The possible strengths and limitations of each method are highlighted, which can guide readers when selecting a method based on a given research question. In particular, we believe that multimodal fusion approaches will shed further light on the neuronal mechanisms underlying the major structural and functional pathophysiological features of both the healthy brain (e.g. development) or the diseased brain (e.g. mental illness) and, in the latter case, may provide a more sensitive measure than unimodal imaging for disease classification, e.g. multimodal biomarkers, which potentially can be used to support clinical diagnosis based on neuroimaging techniques.

Altered Small-World Brain Networks in Schizophrenia Patients during Working Memory Performance

Impairment of working memory (WM) performance in schizophrenia patients (SZ) is well-established. Compared to healthy controls (HC), SZ patients show aberrant blood oxygen level dependent (BOLD) activations and disrupted functional connectivity during WM performance. In this study, we examined the small-world network metrics computed from functional magnetic resonance imaging (fMRI) data collected as 35 HC and 35 SZ performed a Sternberg Item Recognition Paradigm (SIRP) at three WM load levels. Functional connectivity networks were built by calculating the partial correlation on preprocessed time courses of BOLD signal between task-related brain regions of interest (ROIs) defined by group independent component analysis (ICA). The networks were then thresholded within the small-world regime, resulting in undirected binarized small-world networks at different working memory loads. Our results showed: 1) at the medium WM load level, the networks in SZ showed a lower clustering coefficient and less local efficiency compared with HC; 2) in SZ, most network measures altered significantly as the WM load level increased from low to medium and from medium to high, while the network metrics were relatively stable in HC at different WM loads; and 3) the altered structure at medium WM load in SZ was related to their performance during the task, with longer reaction time related to lower clustering coefficient and lower local efficiency. These findings suggest brain connectivity in patients with SZ was more diffuse and less strongly linked locally in functional network at intermediate level of WM when compared to HC. SZ show distinctly inefficient and variable network structures in response to WM load increase, comparing to stable highly clustered network topologies in HC.

Modular Organization of Functional Network Connectivity in Healthy Controls and Patients with Schizophrenia during the Resting State.

Neuroimaging studies have shown that functional brain networks composed from select regions of interest have a modular community structure. However, the organization of functional network connectivity (FNC), comprising a purely data-driven network built from spatially independent brain components, is not yet clear. The aim of this study is to explore the modular organization of FNC in both healthy controls (HCs) and patients with schizophrenia (SZs). Resting state functional magnetic resonance imaging data of HCs and SZs were decomposed into independent components (ICs) by group independent component analysis (ICA). Then weighted brain networks (in which nodes are brain components) were built based on correlations between ICA time courses. Clustering coefficients and connectivity strength of the networks were computed. A dynamic branch cutting algorithm was used to identify modules of the FNC in HCs and SZs. Results show stronger connectivity strength and higher clustering coefficient in HCs with more and smaller modules in SZs. In addition, HCs and SZs had some different hubs. Our findings demonstrate altered modular architecture of the FNC in schizophrenia and provide insights into abnormal topological organization of intrinsic brain networks in this mental illness.