Connectivity dynamics in typical development and its relationship to autistic traits and autism spectrum disorder.

Recent advances in neuroimaging techniques have provided significant insights into developmental trajectories of human brain function. Characterizations of typical neurodevelopment provide a framework for understanding altered neurodevelopment, including differences in brain function related to developmental disorders and psychopathology. Historically, most functional connectivity studies of typical and atypical development operate under the assumption that connectivity remains static over time. We hypothesized that relaxing stationarity assumptions would reveal novel features of both typical brain development related to children on the autism spectrum. We employed a "chronnectomic" (recurring, time-varying patterns of connectivity) approach to evaluate transient states of connectivity using resting-state functional MRI in a population-based sample of 774 6- to 10-year-old children. Dynamic connectivity was evaluated using a sliding-window approach, and revealed four transient states. Internetwork connectivity increased with age in modularized dynamic states, illustrating an important pattern of connectivity in the developing brain. Furthermore, we demonstrated that higher levels of autistic traits and ASD diagnosis were associated with longer dwell times in a globally disconnected state. These results provide a roadmap to the chronnectomic organization of the developing brain and suggest that characteristics of functional brain connectivity are related to children on the autism spectrum.

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

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 Alzheimer's 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), Alzheimer's 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.

Classification of schizophrenia and bipolar patients using static and dynamic resting-state fMRI brain connectivity.

Recently, functional network connectivity (FNC, defined as the temporal correlation among spatially distant brain networks) has been used to examine the functional organization of brain networks in various psychiatric illnesses. Dynamic FNC is a recent extension of the conventional FNC analysis that takes into account FNC changes over short periods of time. While such dynamic FNC measures may be more informative about various aspects of connectivity, there has been no detailed head-to-head comparison of the ability of static and dynamic FNC to perform classification in complex mental illnesses. This paper proposes a framework for automatic classification of schizophrenia, bipolar and healthy subjects based on their static and dynamic FNC features. Also, we compare cross-validated classification performance between static and dynamic FNC. Results show that the dynamic FNC significantly outperforms the static FNC in terms of predictive accuracy, indicating that features from dynamic FNC have distinct advantages over static FNC for classification purposes. Moreover, combining static and dynamic FNC features does not significantly improve the classification performance over the dynamic FNC features alone, suggesting that static FNC does not add any significant information when combined with dynamic FNC for classification purposes. A three-way classification methodology based on static and dynamic FNC features discriminates individual subjects into appropriate diagnostic groups with high accuracy. Our proposed classification framework is potentially applicable to additional mental disorders.

Impact of autocorrelation on functional connectivity.

Although the impact of serial correlation (autocorrelation) in residuals of general linear models for fMRI time-series has been studied extensively, the effect of autocorrelation on functional connectivity studies has been largely neglected until recently. Some recent studies based on results from economics have questioned the conventional estimation of functional connectivity and argue that not correcting for autocorrelation in fMRI time-series results in "spurious" correlation coefficients. In this paper, first we assess the effect of autocorrelation on Pearson correlation coefficient through theoretical approximation and simulation. Then we present this effect on real fMRI data. To our knowledge this is the first work comprehensively investigating the effect of autocorrelation on functional connectivity estimates. Our results show that although FC values are altered, even following correction for autocorrelation, results of hypothesis testing on FC values remain very similar to those before correction. In real data we show this is true for main effects and also for group difference testing between healthy controls and schizophrenia patients. We further discuss model order selection in the context of autoregressive processes, effects of frequency filtering and propose a preprocessing pipeline for connectivity studies.

Functional network connectivity during rest and task conditions: a comparative study.

Functional connectivity (FC) examines temporal statistical dependencies among distant brain regions by means of seed-based analysis or independent component analysis (ICA). Spatial ICA also makes it possible to investigate FC at the network level, termed functional network connectivity (FNC). The dynamics of each network (ICA component), which may consist of several remote regions is described by the ICA time-course of that network; hence, FNC studies statistical dependencies among ICA time-courses. In this article, we compare comprehensively FNC in the resting state and during performance of an auditory oddball (AOD) task in 28 healthy subjects on relevant (nonartifactual) brain networks. The results show global FNC decrease during the performance of the task. In addition, we show that specific networks enlarge and/or demonstrate higher activity during the performance of the task. The results suggest that performing an active task like AOD may be facilitated by recruiting more neurons and higher activation of related networks rather than collaboration among different brain networks. We also evaluated the impact of temporal filtering on FNC analyses. Results showed that the results are not significantly affected by filtering.

Autoconnectivity: A new perspective on human brain function.

BACKGROUND: Autocorrelation (AC) in fMRI time-series is a well-known phenomenon, typically attributed to colored noise and therefore removed from the data. We hypothesize that AC reflects systematic and meaningful signal fluctuations that may be tied to neural activity and provide evidence to support this hypothesis. NEW METHOD: Each fMRI time-series is modeled as an autoregressive process from which the autocorrelation is quantified. Then, autocorrelation during resting-state fMRI and auditory oddball (AOD) task in schizophrenia and healthy volunteers is examined. RESULTS: During resting-state, AC was higher in the visual cortex while during AOD task, frontal part of the brain exhibited higher AC in both groups. AC values were significantly lower in specific brain regions in schizophrenia patients (such as thalamus during resting-state) compared to healthy controls in two independent datasets. Moreover, AC values had significant negative correlation with patients' symptoms. AC differences discriminated patients from healthy controls with high accuracy (resting-state). COMPARISON WITH EXISTING METHODS: Contrary to most prior works, the results suggest AC shows meaningful patterns that are discriminative between patients and controls. Our results are in line with recent works attributing autocorrelation to feedback loop of brain's regulatory circuit. CONCLUSIONS: Autoconnectivity is cognitive state dependent (resting-state vs. task) and mental state dependent (healthy vs. schizophrenia). The concept of autoconnectivity resembles a recurrent neural network and provides a new perspective of functional integration in the brain. These findings may have important implications for understanding of brain function in health and disease as well as for analysis of fMRI time-series.

Using Adipose Measures from Health Care Provider-Based Imaging Data for Discovery.

The location and type of adipose tissue is an important factor in metabolic syndrome. A database of picture archiving and communication system (PACS) derived abdominal computerized tomography (CT) images from a large health care provider, Geisinger, was used for large-scale research of the relationship of volume of subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT) with obesity-related diseases and clinical laboratory measures. Using a "greedy snake" algorithm and 2,545 CT images from the Geisinger PACS, we measured levels of VAT, SAT, total adipose tissue (TAT), and adipose ratio volumes. Sex-combined and sex-stratified association testing was done between adipose measures and 1,233 disease diagnoses and 37 clinical laboratory measures. A genome-wide association study (GWAS) for adipose measures was also performed. SAT was strongly associated with obesity and morbid obesity. VAT levels were strongly associated with type 2 diabetes-related diagnoses (p = 1.5 × 10-58), obstructive sleep apnea (p = 7.7 × 10-37), high-density lipoprotein (HDL) levels (p = 1.42 × 10-36), triglyceride levels (p = 1.44 × 10-43), and white blood cell (WBC) counts (p = 7.37 × 10-9). Sex-stratified tests revealed stronger associations among women, indicating the increased influence of VAT on obesity-related disease outcomes particularly among women. The GWAS identified some suggestive associations. This study supports the utility of pursuing future clinical and genetic discoveries with existing imaging data-derived adipose tissue measures deployed at a larger scale.

Advanced machine learning in action: identification of intracranial hemorrhage on computed tomography scans of the head with clinical workflow integration.

Intracranial hemorrhage (ICH) requires prompt diagnosis to optimize patient outcomes. We hypothesized that machine learning algorithms could automatically analyze computed tomography (CT) of the head, prioritize radiology worklists and reduce time to diagnosis of ICH. 46,583 head CTs (~2 million images) acquired from 2007-2017 were collected from several facilities across Geisinger. A deep convolutional neural network was trained on 37,074 studies and subsequently evaluated on 9499 unseen studies. The predictive model was implemented prospectively for 3 months to re-prioritize "routine" head CT studies as "stat" on realtime radiology worklists if an ICH was detected. Time to diagnosis was compared between the re-prioritized "stat" and "routine" studies. A neuroradiologist blinded to the study reviewed false positive studies to determine whether the dictating radiologist overlooked ICH. The model achieved an area under the ROC curve of 0.846 (0.837-0.856). During implementation, 94 of 347 "routine" studies were re-prioritized to "stat", and 60/94 had ICH identified by the radiologist. Five new cases of ICH were identified, and median time to diagnosis was significantly reduced (p < 0.0001) from 512 to 19 min. In particular, one outpatient with vague symptoms on anti-coagulation was found to have an ICH which was treated promptly with reversal of anticoagulation, resulting in a good clinical outcome. Of the 34 false positives, the blinded over-reader identified four probable ICH cases overlooked in original interpretation. In conclusion, an artificial intelligence algorithm can prioritize radiology worklists to reduce time to diagnosis of new outpatient ICH by 96% and may also identify subtle ICH overlooked by radiologists. This demonstrates the positive impact of advanced machine learning in radiology workflow optimization.

Predicting deterioration of ventricular function in patients with repaired tetralogy of Fallot using machine learning.

Aims: Previous studies using regression analyses have failed to identify which patients with repaired tetralogy of Fallot (rTOF) are at risk for deterioration in ventricular size and function despite using common clinical and cardiac function parameters as well as cardiac mechanics (strain and dyssynchrony). This study used a machine learning pipeline to comprehensively investigate the predictive value of the baseline variables derived from cardiac magnetic resonance (CMR) imaging and provide models for identifying patients at risk for deterioration. Methods and results: Longitudinal deterioration for 153 patients with rTOF was categorized as 'none', 'minor', or 'major' based on changes in ventricular size and ejection fraction between two CMR scans at least 6 months apart (median 2.7 years). Baseline variables were measured at the time of the first CMR. An exhaustive variable search with a support vector machine classifier and five-fold cross-validation was used to predict deterioration and identify the most useful variables. For predicting any deterioration (minor or major) vs. no deterioration, the mean area under the curve (AUC) was 0.82 ± 0.06. For predicting major deterioration vs. minor or no deterioration, the AUC was 0.77 ± 0.07. Baseline left ventricular (LV) ejection fraction, LV circumferential strain, and pulmonary regurgitation were most useful for achieving accurate predictions. Conclusion: For the prediction of deterioration in patients with rTOF, a machine learning pipeline uncovered the utility of baseline variables that was previously lost to regression analyses. The predictive models may be useful for planning early interventions in patients with high risk.

Machine Learning of Functional Magnetic Resonance Imaging Network Connectivity Predicts Substance Abuse Treatment Completion.

BACKGROUND: Successfully treating illicit drug use has become paramount, yet elusive. Devising specialized treatment interventions could increase positive outcomes, but it is necessary to identify risk factors of poor long-term outcomes to develop specialized, efficacious treatments. We investigated whether functional network connectivity (FNC) measures were predictive of substance abuse treatment completion using machine learning pattern classification of functional magnetic resonance imaging data. METHODS: Treatment-seeking stimulant- or heroin-dependent incarcerated participants (n = 139; 89 women) volunteered for a 12-week substance abuse treatment program. Participants performed a response inhibition Go/NoGo functional magnetic resonance imaging task prior to onset of the substance abuse treatment. We tested whether FNC related to the anterior cingulate cortex would be predictive of those who would or would not complete a 12-week substance abuse treatment program. RESULTS: Machine learning pattern classification models using FNC between networks incorporating the anterior cingulate cortex, striatum, and insula predicted which individuals would (sensitivity: 81.31%) or would not (specificity: 78.13%) complete substance abuse treatment. FNC analyses predicted treatment completion above and beyond other clinical assessment measures, including age, sex, IQ, years of substance use, psychopathy, anxiety and depressive symptomatology, and motivation for change. CONCLUSIONS: Aberrant neural network connections predicted substance abuse treatment outcomes, which could illuminate new targets for developing interventions designed to reduce or eliminate substance use while facilitating long-term outcomes. This work represents the first application of machine-learning models of FNC analyses of functional magnetic resonance imaging data to predict which substance abusers would or would not complete treatment.

Sex and Age Effects of Functional Connectivity in Early Adulthood.

Functional connectivity (FC) in resting-state functional magnetic resonance imaging (rs-fMRI) is widely used to find coactivating regions in the human brain. Despite its widespread use, the effects of sex and age on resting FC are not well characterized, especially during early adulthood. Here we apply regression and graph theoretical analyses to explore the effects of sex and age on FC between the 116 AAL atlas parcellations (a total of 6670 FC measures). rs-fMRI data of 494 healthy subjects (203 males and 291 females; age range: 22-36 years) from the Human Connectome Project were analyzed. We report the following findings. (1) Males exhibited greater FC than females in 1352 FC measures (1025 survived Bonferroni correction; [Formula: see text]). In 641 FC measures, females exhibited greater FC than males but none survived Bonferroni correction. Significant FC differences were mainly present in frontal, parietal, and temporal lobes. Although the average FC values for males and females were significantly different, FC values of males and females exhibited large overlap. (2) Age effects were present only in 29 FC measures and all significant age effects showed higher FC in younger subjects. Age and sex differences of FC remained significant after controlling for cognitive measures. (3) Although sex [Formula: see text] age interaction did not survive multiple comparison correction, FC in females exhibited a faster cross-sectional decline with age. (4) Male brains were more locally clustered in all lobes but the cerebellum; female brains had a higher clustering coefficient at the whole-brain level. Our results indicate that although both male and female brains show small-world network characteristics, male brains were more segregated and female brains were more integrated. Findings of this study further our understanding of FC in early adulthood and provide evidence to support that age and sex should be controlled for in FC studies of young adults.

Accurate classification of schizophrenia patients based on novel resting-state fMRI features.

There is a growing interest in automatic classification of mental disorders such as schizophrenia based on neuroimaging data. Most previous studies considered structural MRI, diffusion tensor imaging and task-based fMRI for this purpose. However, resting-state fMRI data has not been used much to evaluate discrimination of schizophrenia patients from healthy controls. Resting data are of great interest, since they are relatively easy to collect, and not confounded by behavioral performance on a task. In this study, we extract two types of features from resting-state fMRI data: functional network connectivity features that capture internetwork connectivity patterns and autoconnectivity features capturing temporal connectivity of each brain network. Autoconnectivity is a novel concept we have recently proposed. We used minimum redundancy maximum relevancy to select features. Classification results using support vector machine shows that combining these two types of features can improve the classification on a large resting fMRI dataset consisting of 195 patients with schizophrenia and 175 healthy controls. We achieved the accuracy of 85% which is very promising.

Detecting volumetric changes in fMRI connectivity networks in schizophrenia patients.

There is a growing interest in identifying neuroimaging-based biomarkers for schizophrenia. Previous studies have shown both functional and structural brain abnormalities in schizophrenia patients. One main category of these findings consists of volumetric abnormalities in brain structure in different cortical and subcortical structures in patients' brain. However there has been little work investigating changes in the brain's functional volumes. Nor has there been work studying differences in brain networks as opposed to single regions. In this study, we investigated the volumes of functional networks as potential biomarkers. Independent component analysis was used to decompose fMRI images into maximally independent spatial maps and corresponding time-courses. Volume of functional networks was computed from subject-specific back reconstructed spatial maps. The results show that different nodes of the default-mode network exhibit volumetric abnormalities in schizophrenia patients. Interestingly these networks are larger in patients compared to controls.

Sharing privacy-sensitive access to neuroimaging and genetics data: a review and preliminary validation.

The growth of data sharing initiatives for neuroimaging and genomics represents an exciting opportunity to confront the "small N" problem that plagues contemporary neuroimaging studies while further understanding the role genetic markers play in the function of the brain. When it is possible, open data sharing provides the most benefits. However, some data cannot be shared at all due to privacy concerns and/or risk of re-identification. Sharing other data sets is hampered by the proliferation of complex data use agreements (DUAs) which preclude truly automated data mining. These DUAs arise because of concerns about the privacy and confidentiality for subjects; though many do permit direct access to data, they often require a cumbersome approval process that can take months. An alternative approach is to only share data derivatives such as statistical summaries-the challenges here are to reformulate computational methods to quantify the privacy risks associated with sharing the results of those computations. For example, a derived map of gray matter is often as identifiable as a fingerprint. Thus alternative approaches to accessing data are needed. This paper reviews the relevant literature on differential privacy, a framework for measuring and tracking privacy loss in these settings, and demonstrates the feasibility of using this framework to calculate statistics on data distributed at many sites while still providing privacy.

Identification of patterns of gray matter abnormalities in schizophrenia using source-based morphometry and bagging.

Despite its reliable diagnosis, schizophrenia lacks an objective diagnostic test or a validated biomarker, which prevents a better understanding of this disorder. Structural magnetic resonance imaging (sMRI) has been vastly explored to find consistent abnormality patterns of gray matter concentration (GMC) in schizophrenia, yet we are far from having reached conclusive evidence. This paper presents a machine learning approach based on resampling techniques to find brain regions with consistent patterns of GMC differences between healthy controls and schizophrenia patients, these regions being detected by means of source-based morphometry. This work uses multi-site data from the Mind Clinical Imaging Consortium, which is composed of sMRI data from 124 controls and 110 patients. Our method achieves a better classification rate than other algorithms and detects regions with GMC differences between both groups that are consistent with several findings on the literature. In addition, the results obtained on data from multiple sites suggest that it may be possible to replicate these results on other datasets.

Brain potentials measured during a Go/NoGo task predict completion of substance abuse treatment.

BACKGROUND: U.S. nationwide estimates indicate that 50% to 80% of prisoners have a history of substance abuse or dependence. Tailoring substance abuse treatment to specific needs of incarcerated individuals could improve effectiveness of treating substance dependence and preventing drug abuse relapse. We tested whether pretreatment neural measures of a response inhibition (Go/NoGo) task would predict which individuals would or would not complete a 12-week cognitive behavioral substance abuse treatment program. METHODS: Adult incarcerated participants (n = 89; women n = 55) who volunteered for substance abuse treatment performed a Go/NoGo task while event-related potentials (ERPs) were recorded. Stimulus- and response-locked ERPs were compared between participants who completed (n = 68; women = 45) and discontinued (n = 21; women = 10) treatment. RESULTS: As predicted, stimulus-locked P2, response-locked error-related negativity (ERN/Ne), and response-locked error positivity (Pe), measured with windowed time-domain and principal component analysis, differed between groups. Using logistic regression and support-vector machine (i.e., pattern classifiers) models, P2 and Pe predicted treatment completion above and beyond other measures (i.e., N2, P300, ERN/Ne, age, sex, IQ, impulsivity, depression, anxiety, motivation for change, and years of drug abuse). CONCLUSIONS: Participants who discontinued treatment exhibited deficiencies in sensory gating, as indexed by smaller P2; error-monitoring, as indexed by smaller ERN/Ne; and adjusting response strategy posterror, as indexed by larger Pe. The combination of P2 and Pe reliably predicted 83.33% of individuals who discontinued treatment. These results may help in the development of individualized therapies, which could lead to more favorable, long-term outcomes.

Classification of schizophrenia patients based on resting-state functional network connectivity.

There is a growing interest in automatic classification of mental disorders based on neuroimaging data. Small training data sets (subjects) and very large amount of high dimensional data make it a challenging task to design robust and accurate classifiers for heterogeneous disorders such as schizophrenia. Most previous studies considered structural MRI, diffusion tensor imaging and task-based fMRI for this purpose. However, resting-state data has been rarely used in discrimination of schizophrenia patients from healthy controls. Resting data are of great interest, since they are relatively easy to collect, and not confounded by behavioral performance on a task. Several linear and non-linear classification methods were trained using a training dataset and evaluate with a separate testing dataset. Results show that classification with high accuracy is achievable using simple non-linear discriminative methods such as k-nearest neighbors (KNNs) which is very promising. We compare and report detailed results of each classifier as well as statistical analysis and evaluation of each single feature. To our knowledge our effects represent the first use of resting-state functional network connectivity (FNC) features to classify schizophrenia.

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.

Functional network connectivity during rest and task: comparison of healthy controls and schizophrenic patients.

Functional connectivity examines temporal statistical dependencies among distant brain regions by means of seed-based analysis or independent component analysis (ICA). Spatial ICA also makes it possible to investigate functional connectivity at the network level, termed functional network connectivity (FNC). The dynamics of each network (ICA component) which may consist of several remote regions is described by the ICA time-course of that network; hence FNC studies statistical dependencies among ICA time-courses. In this paper, we compare comprehensively FNC in the resting state and during performance of an auditory oddball task in 28 healthy subject and 28 schizophrenic patients on relevant (non-artifactual) brain networks. The results show abnormalities both in the resting state and during the task but also the difference of the two states. Moreover, our results suggest that using data both in the resting-state and during the task can better separate the two groups. It is demonstrated that for three pairs of networks, the FNC of the healthy controls resides within a confined region of the correlation space whereas patients behave more sparsely. This can be used to discriminate the two groups based on partitioning the correlation space during the resting state and the task data.