Resting state neurophysiology of agonist-antagonist myoneural interface in persons with transtibial amputation.

The agonist-antagonist myoneural interface (AMI) is an amputation surgery that preserves sensorimotor signaling mechanisms of the central-peripheral nervous systems. Our first neuroimaging study investigating AMI subjects conducted by Srinivasan et al. (2020) focused on task-based neural signatures, and showed evidence of proprioceptive feedback to the central nervous system. The study of resting state neural activity helps non-invasively characterize the neural patterns that prime task response. In this study on resting state functional magnetic resonance imaging in AMI subjects, we compared functional connectivity in patients with transtibial AMI (n = 12) and traditional (n = 7) amputations (TA). To test our hypothesis that we would find significant neurophysiological differences between AMI and TA subjects, we performed a whole-brain exploratory analysis to identify a seed region; namely, we conducted ANOVA, followed by t-test statistics to locate a seed in the salience network. Then, we implemented a seed-based connectivity analysis to gather cluster-level inferences contrasting our subject groups. We show evidence supporting our hypothesis that the AMI surgery induces functional network reorganization resulting in a neural configuration that significantly differs from the neural configuration after TA surgery. AMI subjects show significantly less coupling with regions functionally dedicated to selecting where to focus attention when it comes to salient stimuli. Our findings provide researchers and clinicians with a critical mechanistic understanding of the effect of AMI amputation on brain networks at rest, which has promising implications for improved neurorehabilitation and prosthetic control.

Comparison of hemodynamic response functions obtained from resting-state functional MRI and invasive electrophysiological recordings in rats.

Resting-state functional MRI (rs-fMRI) is a popular technology that has enriched our understanding of brain and spinal cord functioning, including how different regions communicate (connectivity). But fMRI is an indirect measure of neural activity capturing blood hemodynamics. The hemodynamic response function (HRF) interfaces between the unmeasured neural activity and measured fMRI time series. The HRF is variable across brain regions and individuals, and is modulated by non-neural factors. Ignoring this HRF variability causes errors in FC estimates. Hence, it is crucial to reliably estimate the HRF from rs-fMRI data. Robust techniques have emerged to estimate the HRF from fMRI time series. Although such techniques have been validated non-invasively using simulated and empirical fMRI data, thorough invasive validation using simultaneous electrophysiological recordings, the gold standard, has been elusive. This report addresses this gap in the literature by comparing HRFs derived from invasive intracranial electroencephalogram recordings with HRFs estimated from simultaneously acquired fMRI data in six epileptic rats. We found that the HRF shape parameters (HRF amplitude, latency and width) were not significantly different (p>0.05) between ground truth and estimated HRFs. In the single pathological region, the HRF width was marginally significantly different (p=0.03). Our study provides preliminary invasive validation for the efficacy of the HRF estimation technique in reliably estimating the HRF non-invasively from rs-fMRI data directly. This has a notable impact on rs-fMRI connectivity studies, and we recommend that HRF deconvolution be performed to minimize HRF variability and improve connectivity estimates.

The confound of hemodynamic response function variability in human resting-state functional MRI studies.

Functional magnetic resonance imaging (fMRI) is an indirect measure of neural activity with the hemodynamic response function (HRF) coupling it with unmeasured neural activity. The HRF, modulated by several non-neural factors, is variable across brain regions, individuals and populations. Yet, a majority of human resting-state fMRI connectivity studies continue to assume a non-variable HRF. In this article, with supportive prior evidence, we argue that HRF variability cannot be ignored as it substantially confounds within-subject connectivity estimates and between-subjects connectivity group differences. We also discuss its clinical relevance with connectivity impairments confounded by HRF aberrations in several disorders. We present limited data on HRF differences between women and men, which resulted in a 15.4% median error in functional connectivity estimates in a group-level comparison. We also discuss the implications of HRF variability for fMRI studies in the spinal cord. There is a need for more dialogue within the community on the HRF confound, and we hope that our article is a catalyst in the process.

Resting state neurophysiology of agonist-antagonist myoneural interface in persons with transtibial amputation.

The agonist-antagonist myoneural interface (AMI) is a novel amputation surgery that preserves sensorimotor signaling mechanisms of the central-peripheral nervous systems. Our first neuroimaging study investigating AMI subjects (Srinivasan et al., Sci. Transl. Med. 2020) focused on task-based neural signatures, and showed evidence of proprioceptive feedback to the central nervous system. The study of resting state neural activity helps non-invasively characterize the neural patterns that prime task response. In this first study on resting state fMRI in AMI subjects, we compared resting state functional connectivity in patients with transtibial AMI (n=12) and traditional (n=7) amputations, as well as biologically intact control subjects (n=10). We hypothesized that the AMI surgery will induce functional network reorganization that significantly differs from the traditional amputation surgery and also more closely resembles the neural configuration of controls. We found AMI subjects to have lower connectivity with salience and motor seed regions compared to traditional amputees. Additionally, with connections affected in traditional amputees, AMI subjects exhibited a connectivity pattern more closely resembling controls. Lastly, sensorimotor connectivity in amputee cohorts was significantly associated with phantom sensation (R2=0.7, p=0.0008). These findings provide researchers and clinicians with a critical mechanistic understanding of the effects of the AMI surgery on the brain at rest, spearheading future research towards improved prosthetic control and embodiment.

8-channel Tx dipole and 20-channel Rx loop coil array for MRI of the cervical spinal cord at 7 Tesla.

The quality of cervical spinal cord images can be improved by the use of tailored radiofrequency coil solutions for ultra-high field imaging; however, very few commercial and research 7 Tesla radiofrequency coils currently exist for the spinal cord, and in particular those with parallel transmit capabilities. This work presents the design, testing and validation of a pTx/Rx coil for the human neck and cervical/upper-thoracic spinal cord. The pTx portion is composed of 8 dipoles to ensure high homogeneity over this large region of the spinal cord. The Rx portion is made of 20 semi-adaptable overlapping loops to produce high Signal-to-noise ratio (SNR) across the patient population. The coil housing is designed to facilitate patient positioning and comfort, while being tight fitting to ensure high sensitivity. We demonstrate RF shimming capabilities to optimize B 1 + uniformity, power efficiency and/or specific absorption rate (SAR) efficiency. B 1 + homogeneity, SNR and g-factor was evaluated in adult volunteers and demonstrated excellent performance from the occipital lobe down to the T4-T5 level. We compared the proposed coil with two state-of-the-art head and head/neck coils, confirming its superiority in the cervical and upper-thoracic regions of the spinal cord. This coil solution therefore provides a convincing platform for producing the high image quality necessary for clinical and research scanning of the upper spinal cord.

Neural and behavioral effects of modification of visual attention in body dysmorphic disorder.

In individuals with body dysmorphic disorder (BDD), perceptual appearance distortions may be related to selective attention biases and aberrant visual scanning, contributing to imbalances in global vs. detailed visual processing. Treatments for the core symptom of perceptual distortions are underexplored in BDD; yet understanding their mechanistic effects on brain function is critical for rational treatment development. This study tested a behavioral strategy of visual-attention modification on visual system brain connectivity and eye behaviors. We acquired fMRI data in 37 unmedicated adults with BDD and 30 healthy controls. Participants viewed their faces naturalistically (naturalistic viewing), and holding their gaze on the image center (modulated viewing), monitored with an eye-tracking camera. We analyzed dynamic effective connectivity and visual fixation duration. Modulated viewing resulted in longer mean visual fixation duration compared to during naturalistic viewing, across groups. Further, modulated viewing resulted in stronger connectivity from occipital to parietal dorsal visual stream regions, also evident during the subsequent naturalistic viewing, compared with the initial naturalistic viewing, in BDD. Longer fixation duration was associated with a trend for stronger connectivity during modulated viewing. Those with more severe BDD symptoms had weaker dorsal visual stream connectivity during naturalistic viewing, and those with more negative appearance evaluations had weaker connectivity during modulated viewing. In sum, holding a constant gaze on a non-concerning area of one's face may confer increased communication in the occipital/parietal dorsal visual stream, facilitating global/holistic visual processing. This effect shows persistence during subsequent naturalistic viewing. Results have implications for perceptual retraining treatment designs.

Dynamic Effective Connectivity Patterns During Rapid Face Stimuli Presentation in Body Dysmorphic Disorder.

In individuals with body dysmorphic disorder (BDD), perceptual appearance distortions may be related to imbalances in global vs. local visual processing. Understanding the mechanistic brain effects of potential interventions is crucial for rational treatment development. The dorsal visual stream (DVS) is tuned to rapid image presentation, facilitating global/holistic processing, whereas the ventral visual stream (VVS), responsible for local/detailed processing, reduces activation magnitude with shorter stimulus duration. This study tested a strategy of rapid, short-duration face presentation on visual system connectivity. Thirty-eight unmedicated adults with BDD and 29 healthy controls viewed photographs of their faces for short (125 ms, 250 ms, 500 ms) and long (3000 ms) durations during fMRI scan. Dynamic effective connectivity in DVS and VVS was analyzed. BDD individuals exhibited weaker connectivity from occipital to parietal DVS areas than controls for all stimuli durations. Short compared with long viewing durations (125 ms vs. 3,000 ms and 500 ms vs. 3,000 ms) resulted in significantly weaker VVS connectivity from calcarine cortex to inferior occipital gyri in controls; however, there was only a trend for similar results in BDD. The DVS to VVS ratio, representing a balance between global and local processing, incrementally increased with shorter viewing durations in BDD, although it was not statistically significant. In sum, visual systems in those with BDD are not as responsive as in controls to rapid face presentation. Whether rapid face presentation could reduce connectivity in visual systems responsible for local/detailed processing in BDD may necessitate different parameters or strategies. These results provide mechanistic insights for perceptual retraining treatment designs.

Characterization of Hemodynamic Alterations in Schizophrenia and Bipolar Disorder and Their Effect on Resting-State fMRI Functional Connectivity.

Common and distinct neural bases of Schizophrenia (SZ) and bipolar disorder (BP) have been explored using resting-state fMRI (rs-fMRI) functional connectivity (FC). However, fMRI is an indirect measure of neural activity, which is a convolution of the hemodynamic response function (HRF) and latent neural activity. The HRF, which models neurovascular coupling, varies across the brain within and across individuals, and is altered in many psychiatric disorders. Given this background, this study had three aims: quantifying HRF aberrations in SZ and BP, measuring the impact of such HRF aberrations on FC group differences, and exploring the genetic basis of HRF aberrations. We estimated voxel-level HRFs by deconvolving rs-fMRI data obtained from SZ (N = 38), BP (N = 19), and matched healthy controls (N = 35). We identified HRF group differences (P < .05, FDR corrected) in many regions previously implicated in SZ/BP, with mediodorsal, habenular, and central lateral nuclei of the thalamus exhibiting HRF differences in all pairwise group comparisons. Thalamus seed-based FC analysis revealed that ignoring HRF variability results in false-positive and false-negative FC group differences, especially in insula, superior frontal, and lingual gyri. HRF was associated with DRD2 gene expression (P < .05, 1.62 < |Z| < 2.0), as well as with medication dose (P < .05, 1.75 < |Z| < 3.25). In this first study to report HRF aberrations in SZ and BP, we report the possible modulatory effect of dopaminergic signalling on HRF, and the impact that HRF variability can have on FC studies in clinical samples. To mitigate the impact of HRF variability on FC group differences, we suggest deconvolution during data preprocessing.

Magnetic resonance imaging and spectroscopy in late-onset GM2-gangliosidosis.

OBJECTIVE: Our study aimed to quantify structural changes in relation to metabolic abnormalities in the cerebellum, thalamus, and parietal cortex of patients with late-onset GM2-gangliosidosis (LOGG), which encompasses late-onset Tay-Sachs disease (LOTS) and Sandhoff disease (LOSD). METHODS: We enrolled 10 patients with LOGG (7 LOTS, 3 LOSD) who underwent a neurological assessment battery and 7 age-matched controls. Structural MRI and MRS were performed on a 3 T scanner. Structural volumes were obtained from FreeSurfer and normalized by total intracranial volume. Quantified metabolites included N-acetylaspartate (NAA), choline (Cho), myo-inositol (mI), creatine (Cr), and combined glutamate-glutamine (Glx). Metabolic concentrations were corrected for partial volume effects. RESULTS: Structural analyses revealed significant cerebellar atrophy in the LOGG cohort, which was primarily driven by LOTS patients. NAA was lower and mI higher in LOGG, but this was also significantly driven by the LOTS patients. Clinical ataxia deficits (via the Scale for the Assessment and Rating of Ataxia) were associated with neuronal injury (via NAA), neuroinflammation (via mI), and volumetric atrophy in the cerebellum. INTERPRETATION: The decrease of NAA in the cerebellum suggests that, in addition to cerebellar atrophy, there is ongoing impaired neuronal function and/or loss, while an increase in mI indicates possible neuroinflammation in LOGG (more so within the LOTS subvariant). Quantifying cerebellar atrophy in relation to neurometabolic differences in LOGG may lead to improvements in assessing disease severity, progression, and pharmacological efficacy. Lastly, additional neuroimaging studies in LOGG are required to contrast LOTS and LOSD more accurately.

Functional Connectivity-Based Prediction of Autism on Site Harmonized ABIDE Dataset.

OBJECTIVE: The larger sample sizes available from multi-site publicly available neuroimaging data repositories makes machine-learning based diagnostic classification of mental disorders more feasible by alleviating the curse of dimensionality. However, since multi-site data are aggregated post-hoc, i.e. they were acquired from different scanners with different acquisition parameters, non-neural inter-site variability may mask inter-group differences that are at least in part neural in origin. Hence, the advantages gained by the larger sample size in the context of machine-learning based diagnostic classification may not be realized. METHODS: We address this issue using harmonization of multi-site neuroimaging data using the ComBat technique, which is based on an empirical Bayes formulation to remove inter-site differences in data distributions, to improve diagnostic classification accuracy. Specifically, we demonstrate this using ABIDE (Autism Brain Imaging Data Exchange) multi-site data for classifying individuals with Autism from healthy controls using resting state fMRI-based functional connectivity data. RESULTS: Our results show that higher classification accuracies across multiple classification models can be obtained (especially for models based on artificial neural networks) from multi-site data post harmonization with the ComBat technique as compared to without harmonization, outperforming earlier results from existing studies using ABIDE. Furthermore, our network ablation analysis facilitated important insights into autism spectrum disorder pathology and the connectivity in networks shown to be important for classification covaried with verbal communication impairments in Autism. CONCLUSION: Multi-site data harmonization using ComBat improves neuroimaging-based diagnostic classification of mental disorders. SIGNIFICANCE: ComBat has the potential to make AI-based clinical decision-support systems more feasible in psychiatry.

Ultra-high field (7T) functional magnetic resonance imaging in amyotrophic lateral sclerosis: a pilot study.

Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disease of the central nervous system that results in a progressive loss of motor function and ultimately death. It is critical, yet also challenging, to develop non-invasive biomarkers to identify, localize, measure and/or track biological mechanisms implicated in ALS. Such biomarkers may also provide clues to identify potential molecular targets for future therapeutic trials. Herein we report on a pilot study involving twelve participants with ALS and nine age-matched healthy controls who underwent high-resolution resting state functional magnetic resonance imaging at an ultra-high field of 7 Tesla. A group-level whole-brain analysis revealed a disruption in long-range functional connectivity between the superior sensorimotor cortex (in the precentral gyrus) and bilateral cerebellar lobule VI. Post hoc analyses using atlas-derived left and right cerebellar lobule VI revealed decreased functional connectivity in ALS participants that predominantly mapped to bilateral postcentral and precentral gyri. Cerebellar lobule VI is a transition zone between anterior motor networks and posterior non-motor networks in the cerebellum, and is associated with a wide range of key functions including complex motor and cognitive processing tasks. Our observation of the involvement of cerebellar lobule VI adds to the growing number of studies implicating the cerebellum in ALS. Future avenues of scientific investigation should consider how high-resolution imaging at 7T may be leveraged to visualize differences in functional connectivity disturbances in various genotypes and phenotypes of ALS along the ALS-frontotemporal dementia spectrum.

FMRI hemodynamic response function (HRF) as a novel marker of brain function: applications for understanding obsessive-compulsive disorder pathology and treatment response.

The hemodynamic response function (HRF) represents the transfer function linking neural activity with the functional MRI (fMRI) signal, modeling neurovascular coupling. Since HRF is influenced by non-neural factors, to date it has largely been considered as a confound or has been ignored in many analyses. However, underlying biophysics suggests that the HRF may contain meaningful correlates of neural activity, which might be unavailable through conventional fMRI metrics. Here, we estimated the HRF by performing deconvolution on resting-state fMRI data from a longitudinal sample of 25 healthy controls scanned twice and 44 adults with obsessive-compulsive disorder (OCD) before and after 4-weeks of intensive cognitive-behavioral therapy (CBT). HRF response height, time-to-peak and full-width at half-maximum (FWHM) in OCD were abnormal before treatment and normalized after treatment in regions including the caudate. Pre-treatment HRF predicted treatment outcome (OCD symptom reduction) with 86.4% accuracy, using machine learning. Pre-treatment HRF response height in the caudate head and time-to-peak in the caudate tail were top-predictors of treatment response. Time-to-peak in the caudate tail, a region not typically identified in OCD studies using conventional fMRI activation or connectivity measures, may carry novel importance. Additionally, pre-treatment response height in caudate head predicted post-treatment OCD severity (R = -0.48, P = 0.001), and was associated with treatment-related OCD severity changes (R = -0.44, P = 0.0028), underscoring its relevance. With HRF being a reliable marker sensitive to brain function, OCD pathology, and intervention-related changes, these results could guide future studies towards novel discoveries not possible through conventional fMRI approaches like standard BOLD activation or connectivity.

Density-based clustering of static and dynamic functional MRI connectivity features obtained from subjects with cognitive impairment.

Various machine-learning classification techniques have been employed previously to classify brain states in healthy and disease populations using functional magnetic resonance imaging (fMRI). These methods generally use supervised classifiers that are sensitive to outliers and require labeling of training data to generate a predictive model. Density-based clustering, which overcomes these issues, is a popular unsupervised learning approach whose utility for high-dimensional neuroimaging data has not been previously evaluated. Its advantages include insensitivity to outliers and ability to work with unlabeled data. Unlike the popular k-means clustering, the number of clusters need not be specified. In this study, we compare the performance of two popular density-based clustering methods, DBSCAN and OPTICS, in accurately identifying individuals with three stages of cognitive impairment, including Alzheimer's disease. We used static and dynamic functional connectivity features for clustering, which captures the strength and temporal variation of brain connectivity respectively. To assess the robustness of clustering to noise/outliers, we propose a novel method called recursive-clustering using additive-noise (R-CLAN). Results demonstrated that both clustering algorithms were effective, although OPTICS with dynamic connectivity features outperformed in terms of cluster purity (95.46%) and robustness to noise/outliers. This study demonstrates that density-based clustering can accurately and robustly identify diagnostic classes in an unsupervised way using brain connectivity.

MALINI (Machine Learning in NeuroImaging): A MATLAB toolbox for aiding clinical diagnostics using resting-state fMRI data.

Resting-state functional Magnetic Resonance Imaging (rs-fMRI) has been extensively used for diagnostic classification because it does not require task compliance and is easier to pool data from multiple imaging sites, thereby increasing the sample size. A MATLAB-based toolbox called Machine Learning in NeuroImaging (MALINI) for feature extraction and disease classification is presented. The MALINI toolbox extracts functional and effective connectivity features from preprocessed rs-fMRI data and performs classification between healthy and disease groups using any of 18 popular and widely used machine learning algorithms that are based on diverse principles. A consensus classifier combining the power of multiple classifiers is also presented. The utility of the toolbox is illustrated by accompanying data consisting of resting-state functional connectivity features from healthy controls and subjects with various brain-based disorders: autism spectrum disorder from autism brain imaging data exchange (ABIDE), Alzheimer's disease and mild cognitive impairment from Alzheimer's disease neuroimaging initiative (ADNI), attention deficit hyperactivity disorder from ADHD-200, and post-traumatic stress disorder and post-concussion syndrome acquired in-house. Results of classification performed on the above datasets can be obtained from the main article titled "Supervised machine learning for diagnostic classification from large-scale neuroimaging datasets" [1]. The data was divided into homogeneous and heterogeneous splits, such that 80% could be used for training, model building and cross-validation, while the remaining 20% of the data could be used as a hold-out independent test data for replication of the classification performance, to ensure the robustness of the classifiers to population variance in image acquisition site and age of the sample.

Research Trends in the Application of Yoga to Human Health: A Data Science Approach.

Yoga is an integrative mind-body system of wellbeing developed in India since at least three millennia. Yoga has gained considerable attention in recent decades, partly driven by recent research and evidence about its effectiveness. In this work, we extracted research trends on the effects of Yoga on human health from the US National Library of Medicine's PubMed database (peer-reviewed journal papers). We found that Yoga research spans all organ systems and system-wide issues such as pain and cancer. Research on the nervous system far outpaces other systems, which is expected because of the effects of breathing and exercise on stress reduction, which has been a major application of Yoga. The next cluster of impact concerns the musculoskeletal system and pain (both related to the exercise [asana] aspects of Yoga), as well as cardiovascular/endocrine (also related to stress) and cancer. Stress and mental health, pain, diabetes, and cancer are health issues for which a permanent cure is not available in a majority of cases in modern medicine, although alleviating treatments are available. This has probably fueled interest in complementary approaches such as Yoga for these health issues. Research timeline shows that Yoga-related research largely expanded only after the 2000s. There was a specific uptick after 2004. Similar trends are seen if we look at just clinical trials or randomized control trials (RCTs) or systematic reviews. The percentage of trials (Clinical and RCT) among published literature is around 10-15 % This is comparable to other fields that gained traction around 2000s (e.g. non-invasive brain stimulation). Geographical distribution shows that 37% of all Yoga related research output originates in the USA, 19% from India, 13% from Europe and 31% from the rest of the world. Therefore, the interest is widespread and global. At least the uptick in Yoga-related research in the US post-2000s can be attributed to a substantial jump in funding between 1998 and 2005 from US National Institutes of Health's National Center for Complementary and Integrative Health (NCCIH). We can only surmise that research in this field reached a critical mass in late-1990s, which infused more money into this field, generating more research and creating a positive feedback loop that has sustained the growth so far. We propose that in order to sustain or even accelerate future research in the area, rigor and reproducibility must be enhanced in addition to performing more RCT and clinical trials (increasing % of trials to 20-25% from 10-15%). The fruits of research in the field has to reach the common man in terms of evidence-based solutions to health issues. Without this, accelerated funding in democracies such as India and the USA will not be realizable.

DisConICA: a Software Package for Assessing Reproducibility of Brain Networks and their Discriminability across Disorders.

There is a lack of objective biomarkers to accurately identify the underlying etiology and related pathophysiology of disparate brain-based disorders that are less distinguishable clinically. Brain networks derived from resting-state functional magnetic resonance imaging (rs-fMRI) has been a popular tool for discovering candidate biomarkers. Specifically, independent component analysis (ICA) of rs-fMRI data is a powerful multivariate technique for investigating brain networks. However, ICA-derived brain networks that are not highly reproducible within heterogeneous clinical populations may exhibit mean statistical separation between groups, yet not be sufficiently discriminative at the individual-subject level. We hypothesize that functional brain networks that are most reproducible in subjects within clinical and control groups separately, but not when the two groups are merged, may possess the ability to discriminate effectively between the groups even at the individual-subject level. In this study, we present DisConICA or "Discover Confirm Independent Component Analysis", a software package that implements the methodology in support of our hypothesis. It relies on a "discover-confirm" approach based upon the assessment of reproducibility of independent components (representing brain networks) obtained from rs-fMRI (discover phase) using the gRAICAR (generalized Ranking and Averaging Independent Component Analysis by Reproducibility) algorithm, followed by unsupervised clustering analysis of these components to evaluate their ability to discriminate between groups (confirm phase). The unique feature of our software package is its ability to seamlessly interface with other software packages such as SPM and FSL, so that all related analyses utilizing features of other software can be performed within our package, thus providing a one-stop software solution starting with raw DICOM images to the final results. We showcase our software using rs-fMRI data acquired from US Army soldiers returning from the wars in Iraq and Afghanistan who were clinically grouped into the following groups: PTSD (posttraumatic stress disorder), comorbid PCS (post-concussion syndrome) + PTSD, and matched healthy combat controls. This software package along with test data sets is available for download at https://bitbucket.org/masauburn/disconica.

Supervised machine learning for diagnostic classification from large-scale neuroimaging datasets.

There are growing concerns about the generalizability of machine learning classifiers in neuroimaging. In order to evaluate this aspect across relatively large heterogeneous populations, we investigated four disorders: Autism spectrum disorder (N = 988), Attention deficit hyperactivity disorder (N = 930), Post-traumatic stress disorder (N = 87) and Alzheimer's disease (N = 132). We applied 18 different machine learning classifiers (based on diverse principles) wherein the training/validation and the hold-out test data belonged to samples with the same diagnosis but differing in either the age range or the acquisition site. Our results indicate that overfitting can be a huge problem in heterogeneous datasets, especially with fewer samples, leading to inflated measures of accuracy that fail to generalize well to the general clinical population. Further, different classifiers tended to perform well on different datasets. In order to address this, we propose a consensus-classifier by combining the predictive power of all 18 classifiers. The consensus-classifier was less sensitive to unmatched training/validation and holdout test data. Finally, we combined feature importance scores obtained from all classifiers to infer the discriminative ability of connectivity features. The functional connectivity patterns thus identified were robust to the classification algorithm used, age and acquisition site differences, and had diagnostic predictive ability in addition to univariate statistically significant group differences between the groups. A MATLAB toolbox called Machine Learning in NeuroImaging (MALINI), which implements all the 18 different classifiers along with the consensus classifier is available from Lanka et al. (2019) The toolbox can also be found at the following URL: https://github.com/pradlanka/malini .

Deterioration from healthy to mild cognitive impairment and Alzheimer's disease mirrored in corresponding loss of centrality in directed brain networks.

OBJECTIVE: It is important to identify brain-based biomarkers that progressively deteriorate from healthy to mild cognitive impairment (MCI) to Alzheimer's disease (AD). Cortical thickness, amyloid-ß deposition, and graph measures derived from functional connectivity (FC) networks obtained using functional MRI (fMRI) have been previously identified as potential biomarkers. Specifically, in the latter case, betweenness centrality (BC), a nodal graph measure quantifying information flow, is reduced in both AD and MCI. However, all such reports have utilized BC calculated from undirected networks that characterize synchronization rather than information flow, which is better characterized using directed networks. METHODS: Therefore, we estimated BC from directed networks using Granger causality (GC) on resting-state fMRI data (N = 132) to compare the following populations (p < 0.05, FDR corrected for multiple comparisons): normal control (NC), early MCI (EMCI), late MCI (LMCI) and AD. We used an additional metric called middleman power (MP), which not only characterizes nodal information flow as in BC, but also measures nodal power critical for information flow in the entire network. RESULTS: MP detected more brain regions than BC that progressively deteriorated from NC to EMCI to LMCI to AD, as well as exhibited significant associations with behavioral measures. Additionally, graph measures obtained from conventional FC networks could not identify a single node, underscoring the relevance of GC. CONCLUSION: Our findings demonstrate the superiority of MP over BC as well as GC over FC in our case. MP obtained from GC networks could serve as a potential biomarker for progressive deterioration of MCI and AD.

Dynamics of Segregation and Integration in Directional Brain Networks: Illustration in Soldiers With PTSD and Neurotrauma.

Brain functioning relies on various segregated/specialized neural regions functioning as an integrated-interconnected network (i.e., metastability). Various psychiatric and neurologic disorders are associated with aberrant functioning of these brain networks. In this study, we present a novel framework integrating the strength and temporal variability of metastability in brain networks. We demonstrate that this approach provides novel mechanistic insights which enables better imaging-based predictions. Using whole-brain resting-state fMRI and a graph-theoretic framework, we integrated strength and temporal-variability of complex-network properties derived from effective connectivity networks, obtained from 87 U.S. Army soldiers consisting of healthy combat controls (n = 28), posttraumatic stress disorder (PTSD; n = 17), and PTSD with comorbid mild-traumatic brain injury (mTBI; n = 42). We identified prefrontal dysregulation of key subcortical and visual regions in PTSD/mTBI, with all network properties exhibiting lower variability over time, indicative of poorer flexibility. Larger impairment in the prefrontal-subcortical pathway but not prefrontal-visual pathway differentiated comorbid PTSD/mTBI from the PTSD group. Network properties of the prefrontal-subcortical pathway also had significant association (R 2 = 0.56) with symptom severity and neurocognitive performance; and were also found to possess high predictive ability (81.4% accuracy in classifying the disorders, explaining 66-72% variance in symptoms), identified through machine learning. Our framework explained 13% more variance in behaviors compared to the conventional framework. These novel insights and better predictions were made possible by our novel framework using static and time-varying network properties in our three-group scenario, advancing the mechanistic understanding of PTSD and comorbid mTBI. Our contribution has wide-ranging applications for network-level characterization of healthy brains as well as mental disorders.