Human Connectome Project Data Research Articles

The inter-related effects of alcohol use severity and sleep deficiency on semantic processing in young adults

BackgroundBoth alcohol misuse and sleep deficiency are associated with deficits in semantic processing. However, alcohol misuse and sleep deficiency are frequently comorbid and their inter-related effects on semantic processing as well as the underlying neural mechanisms remain to be investigated. MethodsWe curated the Human Connectome Project data of 973 young adults (508 women) to examine the neural correlates of semantic processing in link with the severity of alcohol use and sleep deficiency. The latter were each evaluated using the first principal component (PC1) of principal component analysis of all drinking metrics and the Pittsburgh Sleep Quality Index (PSQI). We employed path modeling to elucidate the interplay among clinical, behavioral, and neural variables. ResultsAmong women, we observed a significant negative correlation between the left precentral gyrus (PCG) and PSQI scores. Mediation analysis revealed that the left PCG activity fully mediated the relationship between PSQI scores and word comprehension in language tasks. In women alone also, the right middle frontal gyrus (MFG) exhibited a significant negative correlation with PC1. The best path model illustrated the associations among PC1, PSQI scores, PCG activity, and MFG activation during semantic processing in women. ConclusionsAlcohol misuse may lead to reduced MFG activation while sleep deficiency hinder semantic processing by suppressing PCG activity in women. The pathway model underscores the influence of sleep quality and alcohol consumption severity on semantic processing in women, suggesting that sex differences in these effects need to be further investigated.

Gray matter volumetric correlates of the polygenic risk of depression: A study of the Human Connectome Project data

Genetic factors confer risks for depression. Understanding the neural endophenotypes, including brain morphometrics, of genetic predisposition to depression would help in unraveling the pathophysiology of depression. We employed voxel-based morphometry (VBM) to examine how gray matter volumes (GMVs) were correlated with the polygenic risk score (PRS) for depression in 993 young adults of the Human Connectome Project. The phenotype of depression was quantified with a DSM-oriented scale of the Achenbach Adult Self-Report. The PRS for depression was computed for each subject using the Psychiatric Genomics Association Study as the base sample. In multiple regression with age, sex, race, drinking severity, and total intracranial volume as covariates, regional GMVs in positive correlation with the PRS were observed in bilateral hippocampi and right gyrus rectus. Regional GMVs in negative correlation with the PRS were observed in a wide swath of brain regions, including bilateral frontal and temporal lobes, anterior cingulate cortex, thalamus, lingual gyri, cerebellum, and the left postcentral gyrus, cuneus, and parahippocampal gyrus. We also found sex difference in anterior cingulate volumes in manifesting the genetic risk of depression. In addition, the GMV of the right cerebellum crus I partially mediated the link from PRS to depression severity. These findings add to the literature by highlighting 1) a more diverse pattern of the volumetric markers of depression, with most regions showing lower but others higher GMVs in association with the genetic risks of depression, and 2) the cerebellar GMV as a genetically informed neural phenotype of depression, in neurotypical individuals.

Functional Networks of Reward and Punishment Processing and Their Molecular Profiles Predicting the Severity of Young Adult Drinking.

Alcohol misuse is associated with altered punishment and reward processing. Here, we investigated neural network responses to reward and punishment and the molecular profiles of the connectivity features predicting alcohol use severity in young adults. We curated the Human Connectome Project data and employed connectome-based predictive modeling (CPM) to examine how functional connectivity (FC) features during wins and losses are associated with alcohol use severity, quantified by Semi-Structured Assessment for the Genetics of Alcoholism, in 981 young adults. We combined the CPM findings and the JuSpace toolbox to characterize the molecular profiles of the network connectivity features of alcohol use severity. The connectomics predicting alcohol use severity appeared specific, comprising less than 0.12% of all features, including medial frontal, motor/sensory, and cerebellum/brainstem networks during punishment processing and medial frontal, fronto-parietal, and motor/sensory networks during reward processing. Spatial correlation analyses showed that these networks were associated predominantly with serotonergic and GABAa signaling. To conclude, a distinct pattern of network connectivity predicted alcohol use severity in young adult drinkers. These "neural fingerprints" elucidate how alcohol misuse impacts the brain and provide evidence of new targets for future intervention.

Discovering the effective connectome of the brain with dynamic Bayesian DAG learning

Understanding the complex mechanisms of the brain can be unraveled by extracting the Dynamic Effective Connectome (DEC). Recently, score-based Directed Acyclic Graph (DAG) discovery methods have shown significant improvements in extracting the causal structure and inferring effective connectivity. However, learning DEC through these methods still faces two main challenges: one with the fundamental impotence of high-dimensional dynamic DAG discovery methods and the other with the low quality of fMRI data. In this paper, we introduce Bayesian Dynamic DAG learning with M-matrices Acyclicity characterization (BDyMA) method to address the challenges in discovering DEC. The presented dynamic DAG enables us to discover direct feedback loop edges as well. Leveraging an unconstrained framework in the BDyMA method leads to more accurate results in detecting high-dimensional networks, achieving sparser outcomes, making it particularly suitable for extracting DEC. Additionally, the score function of the BDyMA method allows the incorporation of prior knowledge into the process of dynamic causal discovery which further enhances the accuracy of results. Comprehensive simulations on synthetic data and experiments on Human Connectome Project (HCP) data demonstrate that our method can handle both of the two main challenges, yielding more accurate and reliable DEC compared to state-of-the-art and traditional methods. Additionally, we investigate the trustworthiness of DTI data as prior knowledge for DEC discovery and show the improvements in DEC discovery when the DTI data is incorporated into the process.

Generative AI for Rapid Diffusion MRI with Improved Image Quality, Reliability and Generalizability

Abstract We use generative AI to enable rapid diffusion MRI (dMRI) with high fidelity, reproducibility, and generalizability across clinical and research settings. We employ a Swin UNEt Transformers (SWIN) model, trained on Human Connectome Project (HCP) data (n=1021) and conditioned on registered T1 scans, to perform generalized dMRI denoising. We also qualitatively demonstrate super-resolution with artificially downsampled HCP data. Remarkably, SWIN can be fine-tuned for an out-of-domain dataset with a single example scan, as we demonstrate on dMRI of children with neurodevelopmental disorders (n=40), adults with acute traumatic brain injury (n=40), and adolescents with intracerebral hemorrhage due to vascular malformations undergoing resection (n = 8), each cohort scanned on different scanner models with different imaging protocols at different sites. This robustness to scan acquisition parameters, patient populations, scanner types, and sites eliminates the advantages of self-supervised methods over our fully-supervised generative AI approach. We exceed current state-of-the-art denoising methods in accuracy and test-retest reliability of rapid diffusion tensor imaging (DTI) requiring only 90 seconds of scan time. SWIN denoising also achieves dramatic improvements over the state-of-the-art for test-retest reliability of intracellular volume fraction and free water fraction measurements and can remove heavy-tail noise, improving biophysical modeling fidelity. SWIN enables rapid diffusion MRI with unprecedented accuracy and reliability, especially at high diffusion-weighting for probing biological tissues at microscopic spatial scales. The code and model are publicly available at https://github.com/ucsfncl/dmri-swin.

DDParcel: Deep Learning Anatomical Brain Parcellation From Diffusion MRI.

Parcellation of anatomically segregated cortical and subcortical brain regions is required in diffusion MRI (dMRI) analysis for region-specific quantification and better anatomical specificity of tractography. Most current dMRI parcellation approaches compute the parcellation from anatomical MRI (T1- or T2-weighted) data, using tools such as FreeSurfer or CAT12, and then register it to the diffusion space. However, the registration is challenging due to image distortions and low resolution of dMRI data, often resulting in mislabeling in the derived brain parcellation. Furthermore, these approaches are not applicable when anatomical MRI data is unavailable. As an alternative we developed the Deep Diffusion Parcellation (DDParcel), a deep learning method for fast and accurate parcellation of brain anatomical regions directly from dMRI data. The input to DDParcel are dMRI parameter maps and the output are labels for 101 anatomical regions corresponding to the FreeSurfer Desikan-Killiany (DK) parcellation. A multi-level fusion network leverages complementary information in the different input maps, at three network levels: input, intermediate layer, and output. DDParcel learns the registration of diffusion features to anatomical MRI from the high-quality Human Connectome Project data. Then, to predict brain parcellation for a new subject, the DDParcel network no longer requires anatomical MRI data but only the dMRI data. Comparing DDParcel's parcellation with T1w-based parcellation shows higher test-retest reproducibility and a higher regional homogeneity, while requiring much less computational time. Generalizability is demonstrated on a range of populations and dMRI acquisition protocols. Utility of DDParcel's parcellation is demonstrated on tractography analysis for fiber tract identification.

Task-induced changes in brain entropy.

Entropy indicates irregularity of a dynamic system, with higher entropy indicating higher irregularity and more transit states. In the human brain, regional brain entropy (BEN) has been increasingly assessed using resting state fMRI (rs-fMRI), while changes of regional BEN during task-based fMRI have been scarcely studied. The purpose of this study is to characterize task-induced regional BEN alterations using the large Human Connectome Project (HCP) data. To control the potential modulation by the block design, BEN of task-fMRI was calculated from the fMRI images acquired during the task conditions only (task BEN) and then compared to BEN of rs-fMRI (resting BEN). Moreover, BEN was separately calculated from the control blocks of the task-fMRI runs (control BEN) and compared to task BEN. Finally, control BEN was compared to resting BEN to test for residual task effects in the control condition. With respect to resting state, task performance unanimously induced BEN reduction in the peripheral cortical area and BEN increase in the centric part of the sensorimotor and perception networks. Control compared to resting BEN showed similar entropy alterations, suggesting large residual task effects. Task compared to control BEN was characterized by reduced entropy in occipital, orbitofrontal, and parietal regions.

Attention-based fusion of multiple graphheat networks for structural to functional brain mapping

Over the last decade, there has been growing interest in learning the mapping from structural connectivity (SC) to functional connectivity (FC) of the brain. The spontaneous fluctuations of the brain activity during the resting-state as captured by functional MRI (rsfMRI) contain rich non-stationary dynamics over a relatively fixed structural connectome. Among the modeling approaches, graph diffusion-based methods with single and multiple diffusion kernels approximating static or dynamic functional connectivity have shown promise in predicting the FC given the SC. However, these methods are computationally expensive, not scalable, and fail to capture the complex dynamics underlying the whole process. Recently, deep learning methods such as GraphHeat networks and graph diffusion have been shown to handle complex relational structures while preserving global information. In this paper, we propose a novel attention-based fusion of multiple GraphHeat networks (A-GHN) for mapping SC-FC. A-GHN enables us to model multiple heat kernel diffusion over the brain graph for approximating the complex Reaction Diffusion phenomenon. We argue that the proposed deep learning method overcomes the scalability and computational inefficiency issues but can still learn the SC-FC mapping successfully. Training and testing were done using the rsfMRI data of 1058 participants from the human connectome project (HCP), and the results establish the viability of the proposed model. On HCP data, we achieve a high Pearson correlation of 0.788 (Desikan-Killiany atlas with 87 regions) and 0.773 (AAL atlas with 86 regions). Furthermore, experiments demonstrate that A-GHN outperforms the existing methods in learning the complex nature of the structure-function relation of the human brain.

Identification and validation of supervariants reveal novel loci associated with human white matter microstructure.

As an essential part of the central nervous system, white matter coordinates communications between different brain regions and is related to a wide range of neurodegenerative and neuropsychiatric disorders. Previous genome-wide association studies (GWASs) have uncovered loci associated with white matter microstructure. However, GWASs suffer from limited reproducibility and difficulties in detecting multi-single-nucleotide polymorphism (multi-SNP) and epistatic effects. In this study, we adopt the concept of supervariants, a combination of alleles in multiple loci, to account for potential multi-SNP effects. We perform supervariant identification and validation to identify loci associated with 22 white matter fractional anisotropy phenotypes derived from diffusion tensor imaging. To increase reproducibility, we use United Kingdom (UK) Biobank White British (n = 30,842) data for discovery and internal validation, and UK Biobank White but non-British (n = 1927) data, Europeans from the Adolescent Brain Cognitive Development study (n = 4399) data, and Europeans from the Human Connectome Project (n = 319) data for external validation. We identify 23 novel loci on the discovery set that have not been reported in the previous GWASs on white matter microstructure. Among them, three supervariants on genomic regions 5q35.1, 8p21.2, and 19q13.32 have P-values lower than 0.05 in the meta-analysis of the three independent validation data sets. These supervariants contain genetic variants located in genes that have been related to brain structures, cognitive functions, and neuropsychiatric diseases. Our findings provide a better understanding of the genetic architecture underlying white matter microstructure.

Structural Connectivity of the Human Piriform Cortex: an Exploratory Study.

The piriform cortex (PC) is part of the primary olfactory network in humans. Recent findings suggest that it plays a role in pathophysiology of epilepsy. Therefore, studying its connectivity can further our understanding of seizure propagation in epilepsy. We aimed to explore the structural connectivity of PC using high-quality human connectome project data coupled with segmentation of PC on anatomic MRI. Twenty subjects were randomly selected from the human connectome project database, and PC was traced on each hemisphere. Probabilistic whole-brain tractography was then used to visualize PC connectivity. The strongest connectivity was noted between PC and ipsilateral insula in both hemispheres. Specifically, the posterior long gyrus of each insula was predominantly connected to PC. This was followed by connections between PC and basal ganglia as well as orbital frontal cortices. The PC has the strongest connectivity with the insula bilaterally. Specifically, the posterior long gyri of insula have the strongest connectivity. This finding may provide additional insight for localizing and treating temporo-insular epilepsy.

Regression models for partially localized fMRI connectivity analyses.

Brain functional connectivity analysis of resting-state functional magnetic resonance imaging (fMRI) data is typically performed in a standardized template space assuming consistency of connections across subjects. Analysis methods can come in the form of one-edge-at-a-time analyses or dimension reduction/decomposition methods. Common to these approaches is an assumption that brain regions are functionally aligned across subjects; however, it is known that this functional alignment assumption is often violated. In this paper, we use subject-level regression models to explain intra-subject variability in connectivity. Covariates can include factors such as geographic distance between two pairs of brain regions, whether the two regions are symmetrically opposite (homotopic), and whether the two regions are members of the same functional network. Additionally, a covariate for each brain region can be included, to account for the possibility that some regions have consistently higher or lower connectivity. This style of analysis allows us to characterize the fraction of variation explained by each type of covariate. Additionally, comparisons across subjects can then be made using the fitted connectivity regression models, offering a more parsimonious alternative to edge-at-a-time approaches. We apply our approach to Human Connectome Project data on 268 regions of interest (ROIs), grouped into eight functional networks. We find that a high proportion of variation is explained by region covariates and network membership covariates, while geographic distance and homotopy have high relative importance after adjusting for the number of predictors. We also find that the degree of data repeatability using our connectivity regression model-which uses only partial location information about pairs of ROI's-is comparably as high as the repeatability obtained using full location information. While our analysis uses data that have been transformed into a common template-space, we also envision the method being useful in multi-atlas registration settings, where subject data remains in its own geometry and templates are warped instead. These results suggest the tantalizing possibility that fMRI connectivity analysis can be performed in subject-space, using less aggressive registration, such as simple affine transformations, multi-atlas subject-space registration, or perhaps even no registration whatsoever.

Sleep Deficits Inter-Link Lower Basal Forebrain-Posterior Cingulate Connectivity and Perceived Stress and Anxiety Bidirectionally in Young Men.

The basal nucleus of Meynert (BNM), a primary source of cholinergic projections to the cortex, plays key roles in regulating the sleep-wake cycle and attention. Sleep deficit is associated with impairment in cognitive and emotional functions. However, whether or how cholinergic circuit, sleep, and cognitive/emotional dysfunction are inter-related remains unclear. We curated the Human Connectome Project data and explored BNM resting state functional connectivities (rsFC) in relation to sleep deficit, based on the Pittsburgh Sleep Quality Index (PSQI), cognitive performance, and subjective reports of emotional states in 687 young adults (342 women). Imaging data were processed with published routines and evaluated at a corrected threshold. We assessed the correlation between BNM rsFC, PSQI, and clinical measurements with Pearson regressions and their inter-relationships with mediation analyses. In whole-brain regressions with age and alcohol use severity as covariates, men showed lower BNM rsFC with the posterior cingulate cortex (PCC) in correlation with PSQI score. No clusters were identified in women at the same threshold. Both BNM-PCC rsFC and PSQI score were significantly correlated with anxiety, perceived stress, and neuroticism scores in men. Moreover, mediation analyses showed that PSQI score mediated the relationship between BNM-PCC rsFC and these measures of negative emotions bidirectionally in men. Sleep deficit is associated with negative emotions and lower BNM rsFC with the PCC. Negative emotional states and BNM-PCC rsFC are bidirectionally related through poor sleep quality. These findings are specific to men, suggesting potential sex differences in the neural circuits regulating sleep and emotional states.

Investigating Habenula Functional Connectivity and Reward-Related Activity in Obesity Using Human Connectome Project Data.

Introduction: The habenula, a brain region involved in aversion, might negatively modulate caloric intake. Functional magnetic resonance imaging (fMRI) studies reported associations between weight loss and habenula functional connectivity. However, whether habenula resting-state functional connectivity (rsFC) and reward-related activity are altered in obesity is yet unknown. Methods: Using data from the Human Connectome Project, we included 300 subjects with various body mass indexes (BMIs) and a healthy long-term blood glucose (hemoglobin A1c [HbA1c]). In addition, we investigated a potential BMI × HbA1c interaction in a separate cohort including subjects with prediabetes (n = 72). Habenula rsFC was assessed using a region of interest (ROI)-to-ROI analysis. Furthermore, a separate analysis using gambling task fMRI data focused on reward-related habenula activity. Results: We did not find an association between BMI and habenula rsFC for any of the ROIs. For the exploratory analysis of the BMI × HbA1c effect, a significant interaction effect was found for the habenula-ventral tegmental area (VTA) connection, but this did not survive multiple comparisons correction. Monetary punishment compared with reward activated the bilateral habenula in the BMI sample, but this activity was not associated with BMI. Discussion: In conclusion, we did not find evidence for an association between BMI and habenula rsFC or reward-related activity. However, there might be an interaction between BMI and HbA1c for the habenula-VTA rsFC, suggestive of a role of the habenula in glucose regulation. Future studies should focus on metabolic parameters in their experimental design to confirm our findings and explore the precise role of the habenula in metabolism.

Brain effective connectome based on fMRI and DTI data: Bayesian causal learning and assessment.

Neuroscientific studies aim to find an accurate and reliable brain Effective Connectome (EC). Although current EC discovery methods have contributed to our understanding of brain organization, their performances are severely constrained by the short sample size and poor temporal resolution of fMRI data, and high dimensionality of the brain connectome. By leveraging the DTI data as prior knowledge, we introduce two Bayesian causal discovery frameworks -the Bayesian GOLEM (BGOLEM) and Bayesian FGES (BFGES) methods- that offer significantly more accurate and reliable ECs and address the shortcomings of the existing causal discovery methods in discovering ECs based on only fMRI data. Moreover, to numerically assess the improvement in the accuracy of ECs with our method on empirical data, we introduce the Pseudo False Discovery Rate (PFDR) as a new computational accuracy metric for causal discovery in the brain. Through a series of simulation studies on synthetic and hybrid data (combining DTI from the Human Connectome Project (HCP) subjects and synthetic fMRI), we demonstrate the effectiveness of our proposed methods and the reliability of the introduced metric in discovering ECs. By employing the PFDR metric, we show that our Bayesian methods lead to significantly more accurate results compared to the traditional methods when applied to the Human Connectome Project (HCP) data. Additionally, we measure the reproducibility of discovered ECs using the Rogers-Tanimoto index for test-retest data and show that our Bayesian methods provide significantly more reliable ECs than traditional methods. Overall, our study's numerical and visual results highlight the potential for these frameworks to significantly advance our understanding of brain functionality.

High-accuracy machine learning techniques for functional connectome fingerprinting and cognitive state decoding.

The human brain is a complex network comprised of functionally and anatomically interconnected brain regions. A growing number of studies have suggested that empirical estimates of brain networks may be useful for discovery of biomarkers of disease and cognitive state. A prerequisite for realizing this aim, however, is that brain networks also serve as reliable markers of an individual. Here, using Human Connectome Project data, we build upon recent studies examining brain-based fingerprints of individual subjects and cognitive states based on cognitively demanding tasks that assess, for example, working memory, theory of mind, and motor function. Our approach achieves accuracy of up to 99% for both identification of the subject of an fMRI scan, and for classification of the cognitive state of a previously unseen subject in a scan. More broadly, we explore the accuracy and reliability of five different machine learning techniques on subject fingerprinting and cognitive state decoding objectives, using functional connectivity data from fMRI scans of a high number of subjects (865) across a number of cognitive states (8). These results represent an advance on existing techniques for functional connectivity-based brain fingerprinting and state decoding. Additionally, 16 different functional connectome (FC) matrix construction pipelines are compared in order to characterize the effects of different aspects of the production of FCs on the accuracy of subject and task classification, and to identify possible confounds.

Relationships Between Brain Intrinsic Connectivity Networks and Measures of Cognition and Emotion: A Study of the Human Connectome Project Data.

Growing evidence suggests that variations in cognitive and emotional behavior are associated with variations in brain function. To achieve a more comprehensive assessment, data-driven techniques, specifically independent component analysis (ICA), can be employed to generate outcome variables that describe unique but complementary aspects of functional connectivity within and between networks. In this study, resting-state fMRI and behavioral data were collected from 50 healthy participants in the Human Connectome Project. The neuropsychological battery evaluated performance in various domains, including episodic memory, fluid intelligence, attention, working memory, executive function, cognitive flexibility, inhibition, and processing speed. Emotional measures were also included to assess emotion recognition and negative affects (sadness, fear, and anger). A multivariate approach was adopted to evaluate the association between cognitive abilities and emotional correlates on spatiotemporal features of intrinsic connectivity networks (ICNs). The results were explored at a false discovery rate-corrected threshold of p < 0.05. There was a significant positive association between within-network connectivity of the left central executive network (CEN) and inhibitory control and attention, and a significant negative association between within-network connectivity of the right CEN and episodic memory. Furthermore, increased within-network connectivity of the default-mode network (DMN) was linked to higher fluid intelligence, while within-network connectivity in the salience network (SN) and dorsal attention network (DAN) was associated with cognitive flexibility. Anger was found to be significantly related to increased functional network connectivity between SN and CEN. Sadness and fear were associated with increased within-network connectivity of the right CEN. Additionally, fear was associated with low-frequency spectral power in SN and DMN. These findings offer new insights into the intricate relation between ICN features and cognitive and emotional functions.

Generalizing prediction of task-evoked brain activity across datasets and populations

Predictions of task-based functional magnetic resonance imaging (fMRI) from task-free resting-state (rs) fMRI have gained popularity over the past decade. This method holds a great promise for studying individual variability in brain function without the need to perform highly demanding tasks. However, in order to be broadly used, prediction models must prove to generalize beyond the dataset they were trained on. In this work, we test the generalizability of prediction of task-fMRI from rs-fMRI across sites, MRI vendors and age-groups. Moreover, we investigate the data requirements for successful prediction. We use the Human Connectome Project (HCP) dataset to explore how different combinations of training sample sizes and number of fMRI datapoints affect prediction success in various cognitive tasks. We then apply models trained on HCP data to predict brain activations in data from a different site, a different MRI vendor (Phillips vs. Siemens scanners) and a different age group (children from the HCP-development project). We demonstrate that, depending on the task, a training set of approximately 20 participants with 100 fMRI timepoints each yields the largest gain in model performance. Nevertheless, further increasing sample size and number of timepoints results in significantly improved predictions, until reaching approximately 450–600 training participants and 800–1000 timepoints. Overall, the number of fMRI timepoints influences prediction success more than the sample size. We further show that models trained on adequate amounts of data successfully generalize across sites, vendors and age groups and provide predictions that are both accurate and individual-specific. These findings suggest that large-scale publicly available datasets may be utilized to study brain function in smaller, unique samples.

Nonparametric group variable selection with multivariate response for connectome-based modelling of cognitive scores

Abstract We study association between the structural connectivity and cognitive profiles using a multi-response nonparametric regression model. The cognitive profiles are quantified by seven cognitive test scores, and structural connectivity is represented by nine nodal attributes of connectivity graphs. These nodal centralities together encode different connectivity profiles in the brain network. Nodal attributes may be grouped together for each node, motivating us to introduce group sparsity for feature selection, and radial basis function (RBF)-nets are used to quantify the regression effects. An efficient computation algorithm is developed. Applying our proposed method to Human Connectome Project data, we obtain several interesting findings related to cognitive functioning.

Diffusion MRI data analysis assisted by deep learning synthesized anatomical images (DeepAnat).

Diffusion MRI is a useful neuroimaging tool for non-invasive mapping of human brain microstructure and structural connections. The analysis of diffusion MRI data often requires brain segmentation, including volumetric segmentation and cerebral cortical surfaces, from additional high-resolution T1-weighted (T1w) anatomical MRI data, which may be unacquired, corrupted by subject motion or hardware failure, or cannot be accurately co-registered to the diffusion data that are not corrected for susceptibility-induced geometric distortion. To address these challenges, this study proposes to synthesize high-quality T1w anatomical images directly from diffusion data using convolutional neural networks (CNNs) (entitled "DeepAnat"), including a U-Net and a hybrid generative adversarial network (GAN), and perform brain segmentation on synthesized T1w images or assist the co-registration using synthesized T1w images. The quantitative and systematic evaluations using data of 60 young subjects provided by the Human Connectome Project (HCP) show that the synthesized T1w images and results for brain segmentation and comprehensive diffusion analysis tasks are highly similar to those from native T1w data. The brain segmentation accuracy is slightly higher for the U-Net than the GAN. The efficacy of DeepAnat is further validated on a larger dataset of 300 more elderly subjects provided by the UK Biobank. Moreover, the U-Nets trained and validated on the HCP and UK Biobank data are shown to be highly generalizable to the diffusion data from Massachusetts General Hospital Connectome Diffusion Microstructure Dataset (MGH CDMD) acquired with different hardware systems and imaging protocols and therefore can be used directly without retraining or with fine-tuning for further improved performance. Finally, it is quantitatively demonstrated that the alignment between native T1w images and diffusion images uncorrected for geometric distortion assisted by synthesized T1w images substantially improves upon that by directly co-registering the diffusion and T1w images using the data of 20 subjects from MGH CDMD. In summary, our study demonstrates the benefits and practical feasibility of DeepAnat for assisting various diffusion MRI data analyses and supports its use in neuroscientific applications.

Neural network algorithms predict new diffusion MRI data for multi-compartmental analysis of brain microstructure in a clinical setting

High angular resolution diffusion imaging (HARDI) is a promising method for advanced analysis of brain microstructure. However, comprehensive HARDI analysis requires multiple acquisitions of diffusion images (multi-shell HARDI), which is time consuming and often impractical in clinical settings. This study aimed to establish neural network models that can predict new diffusion datasets from clinically feasible brain diffusion MRI for multi-shell HARDI. The development included 2 algorithms: multi-layer perceptron (MLP) and convolutional neural network (CNN). Both followed a voxel-based approach for model training (70%), validation (15%), and testing (15%). The investigations involved 2 multi-shell HARDI datasets: 1) 11 healthy subjects from the Human Connectome Project (HCP); and 2) 10 local subjects with multiple sclerosis (MS). To assess outcomes, we conducted neurite orientation dispersion and density imaging using both predicted and original data and compared their orientation dispersion index (ODI) and neurite density index (NDI) in different brain tissues with 2 measures: peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM). Results showed that both models achieved robust predictions, which provided competitive ODI and NDI, especially in brain white matter. The CNN outperformed MLP with the HCP data on both PSNR (p < 0.001) and SSIM (p < 0.01). With the MS data, the models performed similarly. Overall, the optimized neural networks can help generate non-acquired brain diffusion MRI, which will make advanced HARDI analysis possible in clinical practice following further validation. Enabling detailed characterization of brain microstructure will allow enhanced understanding of brain function in both health and disease.