Machine learning in resting-state fMRI analysis

Predicting treatment outcome based on resting-state functional connectivity in internalizing mental disorders: A systematic review and meta-analysis

Predicting treatment outcome in internalizing mental disorders prior to treatment initiation is pivotal for precision mental healthcare. In this regard, resting-state functional connectivity (rs-FC) and machine learning have often shown promising prediction accuracies. This systematic review and meta-analysis evaluates these studies, considering their risk of bias through the Prediction Model Study Risk of Bias Assessment Tool (PROBAST). We examined the predictive performance of features derived from rs-FC, identified features with the highest predictive value, and assessed the employed machine learning pipelines. We searched the electronic databases Scopus, PubMed and PsycINFO on the 12th of December 2022, which resulted in 13 included studies. The mean balanced accuracy for predicting treatment outcome was 77% (95% CI: [72%- 83%]). rs-FC of the dorsolateral prefrontal cortex had high predictive value in most studies. However, a high risk of bias was identified in all studies, compromising interpretability. Methodological recommendations are provided based on a comprehensive exploration of the studies' machine learning pipelines, and potential fruitful developments are discussed.

Beyond network connectivity: A classification approach to brain age prediction with resting-state fMRI

The brain is a complex, dynamic organ that shows differences in the same subject at various periods. Understanding how brain activity changes across age as a function of the brain networks has been greatly abetted by fMRI. Canonical analysis consists of determining how alterations in connectivity patterns (CPs) of certain regions are affected. An alternative approach is taken here by not considering connectivity but rather features computed from recordings at the regions of interest (ROIs). Using machine learning (ML) we assess how neural signals are altered by and prospectively predictive of age and sex via a methodology that is novel in drawing upon pairwise classification across six decades of subjects' chronological ages. ML is used to answer the equally important questions of what properties of the computed features are most predictive as well as which brain networks are most affected by aging. It was found that there is decreased differentiation among the neural signals of older subjects that are separated in age by the same number of years as younger subjects. Furthermore, the burstiness of the signals change at different rates between males and females. The findings provide insight into brain aging via an ROI-based analysis, the consideration of several feature groups, and a novel classification-based ML pipeline. There is also a contribution to understanding the effects of data aggregated from different recording centers on the conclusions of fMRI studies.

Gamma power and beta envelope correlation are potential neural predictors of deep hypnosis

Hypnosis is a psychological intervention that is commonly used to enhance the effectiveness of therapeutic suggestions. Despite extensive fascination and study, the neural mechanisms behind hypnosis remain elusive. In the current study, we undertook a systematic exploration of these neural correlates. We first extracted well-studied neurophysiological features from EEG sensors and source-localized data using spectral analysis and two measures of functional connectivity: weighted phase lag index (wPLI) and power envelope correlation (PEC). Next, we developed classification models that predicted self-rated hypnotic experience based on the extracted feature sets. Our findings reveal that gamma power computed on sensor-level data and beta PEC computed between source-localized brain networks are the top predictors of hypnosis depth. Further, a SHapley Additive exPlanations (SHAP) analysis suggested reduced gamma power in the midline frontal area and increased beta PEC between interhemispheric Dorsal Attention Networks (DAN) contribute to the hypnotic experience. These results broaden our understanding of the neural correlates of deep hypnosis, highlighting potential targets for future research. Moreover, this study demonstrates the potential of using predictive models in understanding the neural underpinnings of self-reported hypnotic depth, offering a template for future investigations.

State-transition dynamics of resting-state functional magnetic resonance imaging data: model comparison and test-to-retest analysis

Electroencephalogram (EEG) microstate analysis entails finding dynamics of quasi-stable and generally recurrent discrete states in multichannel EEG time series data and relating properties of the estimated state-transition dynamics to observables such as cognition and behavior. While microstate analysis has been widely employed to analyze EEG data, its use remains less prevalent in functional magnetic resonance imaging (fMRI) data, largely due to the slower timescale of such data. In the present study, we extend various data clustering methods used in EEG microstate analysis to resting-state fMRI data from healthy humans to extract their state-transition dynamics. We show that the quality of clustering is on par with that for various microstate analyses of EEG data. We then develop a method for examining test-retest reliability of the discrete-state transition dynamics between fMRI sessions and show that the within-participant test-retest reliability is higher than between-participant test-retest reliability for different indices of state-transition dynamics, different networks, and different data sets. This result suggests that state-transition dynamics analysis of fMRI data could discriminate between different individuals and is a promising tool for performing fingerprinting analysis of individuals.

Measures of resting-state brain network segregation and integration vary in relation to data quantity: implications for within and between subject comparisons of functional brain network organization

Measures of functional brain network segregation and integration vary with an individual's age, cognitive ability, and health status. Based on these relationships, these measures are frequently examined to study and quantify large-scale patterns of network organization in both basic and applied research settings. However, there is limited information on the stability and reliability of the network measures as applied to functional time-series; these measurement properties are critical to understand if the measures are to be used for individualized characterization of brain networks. We examine measurement reliability using several human datasets (Midnight Scan Club and Human Connectome Project [both Young Adult and Aging]). These datasets include participants with multiple scanning sessions, and collectively include individuals spanning a broad age range of the adult lifespan. The measurement and reliability of measures of resting-state network segregation and integration vary in relation to data quantity for a given participant's scan session; notably, both properties asymptote when estimated using adequate amounts of clean data. We demonstrate how this source of variability can systematically bias interpretation of differences and changes in brain network organization if appropriate safeguards are not included. These observations have important implications for cross-sectional, longitudinal, and interventional comparisons of functional brain network organization.

Analyzing 20 years of Resting-State fMRI Research: Trends and collaborative networks revealed

Resting-state functional magnetic resonance imaging (rs-fMRI), initially proposed by Biswal et al. in 1995, has emerged as a pivotal facet of neuroimaging research. Its ability to examine brain activity during the resting state without the need for explicit tasks or stimuli has made it an integral component of brain imaging studies. In recent years, rs-fMRI has witnessed substantial growth and found widespread application in the investigation of functional connectivity within the brain. To delineate the developmental trajectory of rs-fMRI over the past two decades, we conducted a comprehensive analysis using bibliometric tool Citespace. Our analysis encompassed publication trends, authorship networks, institutional affiliations, international collaborations, as well as emergent themes in references and keywords. Our study reveals a remarkable increase in the volume of rs-fMRI publications over the past two decades, underscoring the burgeoning interest and potential within this field. Harvard University stands out as the institution with the highest number of research papers published in the realm of RS-fMRI, while the United States holds the highest overall influence in this domain. The recent emergence of keywords such as "machine learning" and "default mode," coupled with citation surges in reference to rs-fMRI, have paved new avenues for research within this field. Our study underscores the critical importance of integrating machine learning techniques into rs-fMRI investigations, offering valuable insights into brain function and disease diagnosis. These findings hold profound significance for the field of neuroscience and may furnish insights for future research employing rs-fMRI as a diagnostic tool for a wide array of neurological disorders, thus emphasizing its pivotal role and potential as a tool for investigating brain functionality.

Preserving specificity in federated graph learning for fMRI-based neurological disorder identification

Resting-state functional magnetic resonance imaging (rs-fMRI) offers a non-invasive approach to examining abnormal brain connectivity associated with brain disorders. Graph neural network (GNN) gains popularity in fMRI representation learning and brain disorder analysis with powerful graph representation capabilities. Training a general GNN often necessitates a large-scale dataset from multiple imaging centers/sites, but centralizing multi-site data generally faces inherent challenges related to data privacy, security, and storage burden. Federated Learning (FL) enables collaborative model training without centralized multi-site fMRI data. Unfortunately, previous FL approaches for fMRI analysis often ignore site-specificity, including demographic factors such as age, gender, and education level. To this end, we propose a specificity-aware federated graph learning (SFGL) framework for rs-fMRI analysis and automated brain disorder identification, with a server and multiple clients/sites for federated model aggregation and prediction. At each client, our model consists of a shared and a personalized branch, where parameters of the shared branch are sent to the server while those of the personalized branch remain local. This can facilitate knowledge sharing among sites and also helps preserve site specificity. In the shared branch, we employ a spatio-temporal attention graph isomorphism network to learn dynamic fMRI representations. In the personalized branch, we integrate vectorized demographic information (i.e., age, gender, and education years) and functional connectivity networks to preserve site-specific characteristics. Representations generated by the two branches are then fused for classification. Experimental results on two fMRI datasets with a total of 1218 subjects suggest that SFGL outperforms several state-of-the-art approaches.

Altered intrinsic brain activity in patients with neuropathic pain after brachial plexus avulsion

OBJECTIVE

This study aimed to investigate brain activity changes in patients suffering from neuropathic pain (NP) following brachial plexus avulsion (BPA).

METHODS

Fifteen patients with NP following BPA and eight healthy participants (HP) were recruited for this study. All participants underwent examination using resting-state functional MRI. The amplitude of low-frequency fluctuation (ALFF) and regional homogeneity (ReHo) were calculated and compared between the BPA group, left-BPA subgroup, right-BPA subgroup, and the HP group using independent samples t-tests.

RESULTS

In the BPA group, there were notable increases in ALFF/ReHo observed in the left rolandic operculum, insula, and supramarginal gyrus, while decreases were observed in the left paracentral lobule, fusiform gyrus, calcarine fissure and surrounding cortex, lingual gyrus, precuneus, as well as the bilateral anterior/median cingulate and paracingulate gyri, supplementary motor area, and cerebellum. In the left-BPA subgroup, elevated ALFF/ReHo levels were identified in the left middle/inferior frontal gyri, rolandic operculum, and supramarginal gyrus, with corresponding decreases in the left calcarine fissure and surrounding cortex, inferior occipital gyrus, fusiform gyrus, lingual gyrus, as well as the bilateral anterior/median cingulate and paracingulate gyri, postcentral gyri, supplementary motor area, paracentral lobules, and cerebellum. The right-BPA subgroup displayed increased ALFF/ReHo in the left frontal lobe, rolandic operculum, insula, fusiform gyrus, and lingual gyrus, as well as the right cerebellum. Conversely, decreases in ALFF/ReHo were observed in the bilateral anterior/median cingulate and paracingulate gyri, calcarine fissure and surrounding cortex, cuneus, and occipital lobes.

CONCLUSIONS

The NP after BPA caused spontaneous activity changes in brain regions associated with linguistic, visual, somatosensory, and motor coordination and processing function. The majority of these abnormal areas were situated in the left cerebral hemisphere, while the effect of cingulate gyri and cerebellum seemed to be bilateral.

Unsupervised contrastive graph learning for resting-state functional MRI analysis and brain disorder detection

Resting-state functional magnetic resonance imaging (rs-fMRI) helps characterize regional interactions that occur in the human brain at a resting state. Existing research often attempts to explore fMRI biomarkers that best predict brain disease progression using machine/deep learning techniques. Previous fMRI studies have shown that learning-based methods usually require a large amount of labeled training data, limiting their utility in clinical practice where annotating data is often time-consuming and labor-intensive. To this end, we propose an unsupervised contrastive graph learning (UCGL) framework for fMRI-based brain disease analysis, in which a pretext model is designed to generate informative fMRI representations using unlabeled training data, followed by model fine-tuning to perform downstream disease identification tasks. Specifically, in the pretext model, we first design a bi-level fMRI augmentation strategy to increase the sample size by augmenting blood-oxygen-level-dependent (BOLD) signals, and then employ two parallel graph convolutional networks for fMRI feature extraction in an unsupervised contrastive learning manner. This pretext model can be optimized on large-scale fMRI datasets, without requiring labeled training data. This model is further fine-tuned on to-be-analyzed fMRI data for downstream disease detection in a task-oriented learning manner. We evaluate the proposed method on three rs-fMRI datasets for cross-site and cross-dataset learning tasks. Experimental results suggest that the UCGL outperforms several state-of-the-art approaches in automated diagnosis of three brain diseases (i.e., major depressive disorder, autism spectrum disorder, and Alzheimer's disease) with rs-fMRI data.

Automatic diagnosis of schizophrenia and attention deficit hyperactivity disorder in rs-fMRI modality using convolutional autoencoder model and interval type-2 fuzzy regression

Nowadays, many people worldwide suffer from brain disorders, and their health is in danger. So far, numerous methods have been proposed for the diagnosis of Schizophrenia (SZ) and attention deficit hyperactivity disorder (ADHD), among which functional magnetic resonance imaging (fMRI) modalities are known as a popular method among physicians. This paper presents an SZ and ADHD intelligent detection method of resting-state fMRI (rs-fMRI) modality using a new deep learning method. The University of California Los Angeles dataset, which contains the rs-fMRI modalities of SZ and ADHD patients, has been used for experiments. The FMRIB software library toolbox first performed preprocessing on rs-fMRI data. Then, a convolutional Autoencoder model with the proposed number of layers is used to extract features from rs-fMRI data. In the classification step, a new fuzzy method called interval type-2 fuzzy regression (IT2FR) is introduced and then optimized by genetic algorithm, particle swarm optimization, and gray wolf optimization (GWO) techniques. Also, the results of IT2FR methods are compared with multilayer perceptron, k-nearest neighbors, support vector machine, random forest, and decision tree, and adaptive neuro-fuzzy inference system methods. The experiment results show that the IT2FR method with the GWO optimization algorithm has achieved satisfactory results compared to other classifier methods. Finally, the proposed classification technique was able to provide 72.71% accuracy.

Predicting the efficacy of non-steroidal anti-inflammatory drugs in migraine using deep learning and three-dimensional T1-weighted images

Deep learning (DL) models based on individual images could contribute to tailored therapies and personalized treatment strategies. We aimed to construct a DL model using individual 3D structural images for predicting the efficacy of non-steroidal anti-inflammatory drugs (NSAIDs) in migraine. A 3D convolutional neural network model was constructed, with ResNet18 as the classification backbone, to link structural images to predict the efficacy of NSAIDs. In total, 111 patients were included and allocated to the training and testing sets in a 4:1 ratio. The prediction accuracies of the ResNet34, ResNet50, ResNeXt50, DenseNet121, and 3D ResNet18 models were 0.65, 0.74, 0.65, 0.70, and 0.78, respectively. This model, based on individual 3D structural images, demonstrated better predictive performance in comparison to conventional models. Our study highlights the feasibility of the DL algorithm based on brain structural images and suggests that it can be applied to predict the efficacy of NSAIDs in migraine treatment.

Characterization of Resting-State Functional Connectivity Changes in Hypertension by a Modified Difference Degree Test

Introduction: Hypertension affects over a billion people worldwide, and the application of neuroimaging may elucidate changes brought about by the disease. We have applied a graph theory approach to examine the organizational differences in resting-state functional magnetic resonance imaging (rs-fMRI) data between hypertensive and normotensive participants. To detect these groupwise differences, we performed statistical testing using a modified difference degree test (DDT). Methods: Structural and rs-fMRI data were collected from a cohort of 52 total (29 hypertensive and 23 normotensive) participants. Functional connectivity maps were obtained by partial correlation analysis of participant rs-fMRI data. We modified the DDT null generation algorithm and validated the change through different simulation schemes and then applied this modified DDT to our experimental data. Results: Through a comparative analysis, the modified DDT showed higher true positivity rates (TPR) when compared with the base DDT while also maintaining false positivity rates below the nominal value of 5% in nearly all analytically thresholded trials. Applying the modified DDT to our rs-fMRI data showed differential organization in the hypertension group in the regions throughout the brain including the default mode network. These experimental findings agree with previous studies. Conclusions: While our findings agree with previous studies, the experimental results presented require more investigation to prove their link to hypertension. Meanwhile, our modification to the DDT results in higher accuracy and an increased ability to discern groupwise differences in rs-fMRI data. We expect this to be useful in studying groupwise organizational differences in future studies.

Identifying biomarkers for tDCS treatment response in Alzheimer's disease patients: a machine learning approach using resting-state EEG classification

Background

Transcranial direct current stimulation (tDCS) is a promising treatment for Alzheimer's Disease (AD). However, identifying objective biomarkers that can predict brain stimulation efficacy, remains a challenge. The primary aim of this investigation is to delineate the cerebral regions implicated in AD, taking into account the existing lacuna in comprehension of these regions. In pursuit of this objective, we have employed a supervised machine learning algorithm to prognosticate the neurophysiological outcomes resultant from the confluence of tDCS therapy plus cognitive intervention within both the cohort of responders and non-responders to antecedent tDCS treatment, stratified on the basis of antecedent cognitive outcomes.

Methods

The data were obtained through an interventional trial. The study recorded high-resolution electroencephalography (EEG) in 70 AD patients and analyzed spectral power density during a 6 min resting period with eyes open focusing on a fixed point. The cognitive response was assessed using the AD Assessment Scale-Cognitive Subscale. The training process was carried out through a Random Forest classifier, and the dataset was partitioned into K equally-partitioned subsamples. The model was iterated k times using K-1 subsamples as the training bench and the remaining subsample as validation data for testing the model.

Results

A clinical discriminating EEG biomarkers (features) was found. The ML model identified four brain regions that best predict the response to tDCS associated with cognitive intervention in AD patients. These regions included the channels: FC1, F8, CP5, Oz, and F7.

Conclusion

These findings suggest that resting-state EEG features can provide valuable information on the likelihood of cognitive response to tDCS plus cognitive intervention in AD patients. The identified brain regions may serve as potential biomarkers for predicting treatment response and maybe guide a patient-centered strategy.

Clinical Trial Registration

https://classic.clinicaltrials.gov/ct2/show/NCT02772185?term=NCT02772185&draw=2&rank=1, identifier ID: NCT02772185.

The diagnostic performance of machine learning based on resting-state functional magnetic resonance imaging data for major depressive disorders: a systematic review and meta-analysis

Objective

Machine learning (ML) has been widely used to detect and evaluate major depressive disorder (MDD) using neuroimaging data, i.e., resting-state functional magnetic resonance imaging (rs-fMRI). However, the diagnostic efficiency is unknown. The aim of the study is to conduct an updated meta-analysis to evaluate the diagnostic performance of ML based on rs-fMRI data for MDD.

Methods

English databases were searched for relevant studies. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) was used to assess the methodological quality of the included studies. A random-effects meta-analytic model was implemented to investigate the diagnostic efficiency, including sensitivity, specificity, diagnostic odds ratio (DOR), and area under the curve (AUC). Regression meta-analysis and subgroup analysis were performed to investigate the cause of heterogeneity.

Results

Thirty-one studies were included in this meta-analysis. The pooled sensitivity, specificity, DOR, and AUC with 95% confidence intervals were 0.80 (0.75, 0.83), 0.83 (0.74, 0.82), 14.00 (9, 22.00), and 0.86 (0.83, 0.89), respectively. Substantial heterogeneity was observed among the studies included. The meta-regression showed that the leave-one-out cross-validation (loocv) (sensitivity: p < 0.01, specificity: p < 0.001), graph theory (sensitivity: p < 0.05, specificity: p < 0.01), n > 100 (sensitivity: p < 0.001, specificity: p < 0.001), simens equipment (sensitivity: p < 0.01, specificity: p < 0.001), 3.0T field strength (Sensitivity: p < 0.001, specificity: p = 0.04), and Beck Depression Inventory (BDI) (sensitivity: p = 0.04, specificity: p = 0.06) might be the sources of heterogeneity. Furthermore, the subgroup analysis showed that the sample size (n > 100: sensitivity: 0.71, specificity: 0.72, n < 100: sensitivity: 0.81, specificity: 0.79), the different levels of disease evaluated by the Hamilton Depression Rating Scale (HDRS/HAMD) (mild vs. moderate vs. severe: sensitivity: 0.52 vs. 0.86 vs. 0.89, specificity: 0.62 vs. 0.78 vs. 0.82, respectively), the depression scales in patients with comparable levels of severity. (BDI vs. HDRS/HAMD: sensitivity: 0.86 vs. 0.87, specificity: 0.78 vs. 0.80, respectively), and the features (graph vs. functional connectivity: sensitivity: 0.84 vs. 0.86, specificity: 0.76 vs. 0.78, respectively) selected might be the causes of heterogeneity.

Conclusion

ML showed high accuracy for the automatic diagnosis of MDD. Future studies are warranted to promote the potential use of these classification algorithms in clinical settings.

A low dimensional embedding of brain dynamics enhances diagnostic accuracy and behavioral prediction in stroke

Large-scale brain networks reveal structural connections as well as functional synchronization between distinct regions of the brain. The latter, referred to as functional connectivity (FC), can be derived from neuroimaging techniques such as functional magnetic resonance imaging (fMRI). FC studies have shown that brain networks are severely disrupted by stroke. However, since FC data are usually large and high-dimensional, extracting clinically useful information from this vast amount of data is still a great challenge, and our understanding of the functional consequences of stroke remains limited. Here, we propose a dimensionality reduction approach to simplify the analysis of this complex neural data. By using autoencoders, we find a low-dimensional representation encoding the fMRI data which preserves the typical FC anomalies known to be present in stroke patients. By employing the latent representations emerging from the autoencoders, we enhanced patients' diagnostics and severity classification. Furthermore, we showed how low-dimensional representation increased the accuracy of recovery prediction.

Explaining Deep Learning-Based Representations of Resting State Functional Connectivity Data: Focusing on Interpreting Nonlinear Patterns in Autism Spectrum Disorder

Background

Resting state Functional Magnetic Resonance Imaging fMRI (rs-fMRI) has been used to study brain function in psychiatric disorders, yielding insight into brain organization. However, the high dimensionality of the rs-fMRI data presents challenges, and requires dimensionality reduction before applying machine learning techniques. Neural networks, specifically variational autoencoders (VAEs), have been instrumental in extracting low-dimensional latent representations of resting state functional connectivity patterns, addressing the complex nonlinear structure of rs-fMRI. However, interpreting those latent representations remains a challenge. This paper aims to address this gap by creating explainable VAE models and testing their utility using rs-fMRI data in autism spectrum disorder (ASD).

Methods

One-thousand one hundred and fifty participants (601 HC and 549 patients with ASD) were included in the analysis. We extracted functional connectivity correlation matrices from the preprocessed rs-fMRI data using Power atlas with 264 ROIs. Then VAEs were trained in an unsupervised fashion. Lastly, we introduce our latent contribution scores to explain the relationship between estimated representations and the original rs-fMRI brain measures.

Results

We quantified the latent contribution scores for the ASD and control groups at the network level. We found that both ASD and control groups share the top network connectivity that contribute to all estimated latent components. For example, latent 0 was driven by resting state functional connectivity patterns (rsFC) within ventral attention network in both the ASD and control. However, significant differences in the latent contribution scores between the ASD and control groups were discovered within the ventral attention network in latent 0 and the sensory/somatomotor network in latent 2.

Conclusion

This study introduced latent contribution scores to interpret nonlinear patterns identified by VAEs. These scores effectively capture changes in each observed rsFC features as estimated latent representation changes, enabling an explainable deep learning model to better understand the underlying neural mechanism of ASD.

Multi-Scale Dynamic Graph Learning for Brain Disorder Detection With Functional MRI

Resting-state functional magnetic resonance imaging (rs-fMRI) has been widely used in the detection of brain disorders such as autism spectrum disorder based on various machine/deep learning techniques. Learning-based methods typically rely on functional connectivity networks (FCNs) derived from blood-oxygen-level-dependent time series of rs-fMRI data to capture interactions between brain regions-of-interest (ROIs). Graph neural networks have been recently used to extract fMRI features from graph-structured FCNs, but cannot effectively characterize spatiotemporal dynamics of FCNs, e.g., the functional connectivity of brain ROIs is dynamically changing in a short period of time. Also, many studies usually focus on single-scale topology of FCN, thereby ignoring the potential complementary topological information of FCN at different spatial resolutions. To this end, in this paper, we propose a multi-scale dynamic graph learning (MDGL) framework to capture multi-scale spatiotemporal dynamic representations of rs-fMRI data for automated brain disorder diagnosis. The MDGL framework consists of three major components: 1) multi-scale dynamic FCN construction using multiple brain atlases to model multi-scale topological information, 2) multi-scale dynamic graph representation learning to capture spatiotemporal information conveyed in fMRI data, and 3) multi-scale feature fusion and classification. Experimental results on two datasets show that MDGL outperforms several state-of-the-art methods.

Functional MRI in Neuro-Oncology: State of the Art and Future Directions

Since its discovery in the early 1990s, functional MRI (fMRI) has been used to study human brain function. One well-established application of fMRI in the clinical setting is the neurosurgical planning of patients with brain tumors near eloquent cortical areas. Clinical fMRI aims to preoperatively identify eloquent cortices that serve essential functions in daily life, such as hand movement and language. The primary goal of neurosurgery is to maximize tumor resection while sparing eloquent cortices adjacent to the tumor. When a lesion presents in the vicinity of an eloquent cortex, surgeons may use fMRI to plan their best surgical approach by determining the proximity of the lesion to regions of activation, providing guidance for awake brain surgery and intraoperative brain mapping. The acquisition of fMRI requires patient preparation prior to imaging, determination of functional paradigms, monitoring of patient performance, and both processing and analysis of images. Interpretation of fMRI maps requires a strong understanding of functional neuroanatomy and familiarity with the technical limitations frequently present in brain tumor imaging, including neurovascular uncoupling, patient compliance, and data analysis. This review discusses clinical fMRI in neuro-oncology, relevant ongoing research topics, and prospective future developments in this exciting discipline.

Understanding and predicting future relapse in depression from resting state functional connectivity and self-referential processing

BACKGROUND

The recurrent nature of Major Depressive Disorder (MDD) asks for a better understanding of mechanisms underlying relapse. Previously, self-referential processing abnormalities have been linked to vulnerability for relapse. We investigated whether abnormalities in self-referential cognitions and functioning of associated brain-networks persist upon remission and predict relapse.

METHODS

Remitted recurrent MDD patients (n = 48) and never-depressed controls (n = 23) underwent resting-state fMRI scanning at baseline and were additionally assessed for their implicit depressed self-associations and ruminative behaviour. A template-based dual regression approach was used to investigate between-group differences in default mode, cingulo-opercular and frontoparietal network resting-state functional connectivity (RSFC). Additional prediction of relapse status at 18-month follow-up was investigated within patients using both regression analyses and machine learning classifiers.

RESULTS

Remitted patients showed higher rumination, but no implicit depressed self-associations or RSFC abnormalities were observed between patients and controls. Nevertheless, relapse was related to i) baseline RSFC between the ventral default mode network and the precuneus, dorsomedial frontal gyrus, and inferior occipital lobe, ii) implicit self-associations, and iii) uncontrollability of ruminative thinking, when controlled for depressive symptomatology. Moreover, preliminary machine learning classifiers demonstrated that RSFC within the investigated networks predicted relapse on an individual basis.

CONCLUSIONS

Remitted MDD patients seem to be commonly characterized by abnormal rumination, but not by implicit self-associations or abnormalities in relevant brain networks. Nevertheless, relapse was predicted by self-related cognitions and default mode RSFC during remission, suggesting that variations in self-relevant processing play a role in the complex dynamics associated with the vulnerability to developing recurrent depressive episodes.

CLINICAL TRIAL REGISTRATION

Netherlands Trial Register, August 18, 2015, trial number NL53205.042.15.

A Novel Neurorehabilitation Prognosis Prediction Modeling on Separated Left-Right Hemiplegia Based on Brain-Computer Interfaces Assisted Rehabilitation

It is essential for neuroscience and clinic to estimate the influence of neuro-intervention after brain damage. Most related studies have used Mirrored Contralesional-Ipsilesional hemispheres (MCI) methods flipping the axial neuroimaging on the x-axis in prognosis prediction. But left-right hemispheric asymmetry in the brain has become a consensus. MCI confounds the intrinsic brain asymmetry with the asymmetry caused by unilateral damage, leading to questions about the reliability of the results and difficulties in physiological explanations. We proposed the Separated Left-Right hemiplegia (SLR) method to model left and right hemiplegia separately. Two pipelines have been designed in contradistinction to demonstrate the validity of the SLR method, including MCI and removing intrinsic asymmetry (RIA) pipelines. A patient dataset with 18 left-hemiplegic and 22 right-hemiplegic stroke patients and a healthy dataset with 40 subjects, age- and sex-matched with the patients, were selected in the experiment. Blood-Oxygen Level-Dependent MRI and Diffusion Tensor Imaging were used to build brain networks whose nodes were defined by the Automated Anatomical Labeling atlas. We applied the same statistical and machine learning framework for all pipelines, logistic regression, artificial neural network, and support vector machine for classifying the patients who are significant or non-significant responders to brain-computer interfaces assisted training and optimal subset regression, support vector regression for predicting post-intervention outcomes. The SLR pipeline showed 5-15% improvement in accuracy and at least 0.1 upgrades in [Formula: see text], revealing common and unique recovery mechanisms after left and right strokes and helping clinicians make rehabilitation plans.

Subject-specific whole-brain parcellations of nodes and boundaries are modulated differently under 10 Hz rTMS

Repetitive transcranial magnetic stimulation (rTMS) has gained considerable importance in the treatment of neuropsychiatric disorders, including major depression. However, it is not yet understood how rTMS alters brain's functional connectivity. Here we report changes in functional connectivity captured by resting state functional magnetic resonance imaging (rsfMRI) within the first hour after 10 Hz rTMS. We apply subject-specific parcellation schemes to detect changes (1) in network nodes, where the strongest functional connectivity of regions is observed, and (2) in network boundaries, where functional transitions between regions occur. We use support vector machine (SVM), a widely used machine learning algorithm that is robust and effective, for the classification and characterization of time intervals of changes in node and boundary maps. Our results reveal that changes in connectivity at the boundaries are slower and more complex than in those observed in the nodes, but of similar magnitude according to accuracy confidence intervals. These results were strongest in the posterior cingulate cortex and precuneus. As network boundaries are indeed under-investigated in comparison to nodes in connectomics research, our results highlight their contribution to functional adjustments to rTMS.

Graph Self-Supervised Learning With Application to Brain Networks Analysis

The less training data and insufficient supervision limit the performance of the deep supervised models for brain disease diagnosis. It is significant to construct a learning framework that can capture more information in limited data and insufficient supervision. To address these issues, we focus on self-supervised learning and aim to generalize the self-supervised learning to the brain networks, which are non-Euclidean graph data. More specifically, we propose an ensemble masked graph self-supervised framework named BrainGSLs, which incorporates 1) a local topological-aware encoder that takes the partially visible nodes as input and learns these latent representations, 2) a node-edge bi-decoder that reconstructs the masked edges by the representations of both the masked and visible nodes, 3) a signal representation learning module for capturing temporal representations from BOLD signals and 4) a classifier used for the classification. We evaluate our model on three real medical clinical applications: diagnosis of Autism Spectrum Disorder (ASD), diagnosis of Bipolar Disorder (BD) and diagnosis of Major Depressive Disorder (MDD). The results suggest that the proposed self-supervised training has led to remarkable improvement and outperforms state-of-the-art methods. Moreover, our method is able to identify the biomarkers associated with the diseases, which is consistent with the previous studies. We also explore the correlation of these three diseases and find the strong association between ASD and BD. To the best of our knowledge, our work is the first attempt of applying the idea of self-supervised learning with masked autoencoder on the brain network analysis.

Fused Brain Functional Connectivity Network and Edge-attention Graph Convolution Network for Fibromyalgia Syndrome Diagnosis

Fibromyalgia syndrome (FMS) is a type of rheumatology that seriously affects the normal life of patients. Due to the complex clinical manifestations of FMS, it is challenging to detect FMS. Therefore, an automatic FMS diagnosis model is urgently needed to assist physicians. Brain functional connectivity networks (BFCNs) constructed by resting-state functional magnetic resonance imaging (rs-fMRI) to describe brain functions have been widely used to identify individuals with relevant diseases from normal control (NC). Therefore, we propose a novel model based on BFCN and graph convolutional network (GCN) for automatic FMS diagnosis. Firstly, a novel fused BFCN method is proposed by fusing Pearson's correlation (PC) and low-rank (LR) BFCN, which retains information and reduces data redundancy to construct BFCN. Then we combine the feature of BFCN with non-image information of subjects to obtain nodes and adjacency matrices, which builds a graph with edge attention. Finally, the graph is sent to the GCN layer for FMS diagnosis. Our model is evaluated on the in-house FMS dataset to achieve 82.48% accuracy. The experimental results show that our method outperforms the state-of-the-art competing methods.

Applying Joint Graph Embedding to Study Alzheimer's Neurodegeneration Patterns in Volumetric Data

Neurodegeneration measured through volumetry in MRI is recognized as a potential Alzheimer's Disease (AD) biomarker, but its utility is limited by lack of specificity. Quantifying spatial patterns of neurodegeneration on a whole brain scale rather than locally may help address this. In this work, we turn to network based analyses and extend a graph embedding algorithm to study morphometric connectivity from volume-change correlations measured with structural MRI on the timescale of years. We model our data with the multiple random eigengraphs framework, as well as modify and implement a multigraph embedding algorithm proposed earlier to estimate a low dimensional embedding of the networks. Our version of the algorithm guarantees meaningful finite-sample results and estimates maximum likelihood edge probabilities from population-specific network modes and subject-specific loadings. Furthermore, we propose and implement a novel statistical testing procedure to analyze group differences after accounting for confounders and locate significant structures during AD neurodegeneration. Family-wise error rate is controlled at 5% using permutation testing on the maximum statistic. We show that results from our analysis reveal networks dominated by known structures associated to AD neurodegeneration, indicating the framework has promise for studying AD. Furthermore, we find network-structure tuples that are not found with traditional methods in the field.

Using histogram analysis of the intrinsic brain activity mapping to identify essential tremor

Background

Essential tremor (ET) is one of the most common movement disorders. Histogram analysis based on brain intrinsic activity imaging is a promising way to identify ET patients from healthy controls (HCs) and further explore the spontaneous brain activity change mechanisms and build the potential diagnostic biomarker in ET patients.

Methods

The histogram features based on the Resting-state functional magnetic resonance imaging (Rs-fMRI) data were extracted from 133 ET patients and 135 well-matched HCs as the input features. Then, a two-sample t-test, the mutual information, and the least absolute shrinkage and selection operator methods were applied to reduce the feature dimensionality. Support vector machine (SVM), logistic regression (LR), random forest (RF), and k-nearest neighbor (KNN) were used to differentiate ET and HCs, and classification performance of the established models was evaluated by the mean area under the curve (AUC). Moreover, correlation analysis was carried out between the selected histogram features and clinical tremor characteristics.

Results

Each classifier achieved a good classification performance in training and testing sets. The mean accuracy and area under the curve (AUC) of SVM, LR, RF, and KNN in the testing set were 92.62%, 0.948; 92.01%, 0.942; 93.88%, 0.941; and 92.27%, 0.939, respectively. The most power-discriminative features were mainly located in the cerebello-thalamo-motor and non-motor cortical pathways. Correlation analysis showed that there were two histogram features negatively and one positively correlated with tremor severity.

Conclusion

Our findings demonstrated that the histogram analysis of the amplitude of low-frequency fluctuation (ALFF) images with multiple machine learning algorithms could identify ET patients from HCs and help to understand the spontaneous brain activity pathogenesis mechanisms in ET patients.

Functional alteration due to structural damage is network dependent: insight from multiple sclerosis

Little is known about how the brain's functional organization changes over time with respect to structural damage. Using multiple sclerosis as a model of structural damage, we assessed how much functional connectivity (FC) changed within and between preselected resting-state networks (RSNs) in 122 subjects (72 with multiple sclerosis and 50 healthy controls). We acquired the structural, diffusion, and functional MRI to compute functional connectomes and structural disconnectivity profiles. Change in FC was calculated by comparing each multiple sclerosis participant's pairwise FC to controls, while structural disruption (SD) was computed from abnormalities in diffusion MRI via the Network Modification tool. We used an ordinary least squares regression to predict the change in FC from SD for 9 common RSNs. We found clear differences in how RSNs functionally respond to structural damage, namely that higher-order networks were more likely to experience changes in FC in response to structural damage (default mode R2 = 0.160-0.207, P < 0.001) than lower-order sensory networks (visual network 1 R2 = 0.001-0.007, P = 0.157-0.387). Our findings suggest that functional adaptability to structural damage depends on how involved the affected network is in higher-order processing.

One Size Does Not Fit All: Methodological Considerations for Brain-Based Predictive Modeling in Psychiatry

Psychiatric illnesses are heterogeneous in nature. No illness manifests in the same way across individuals, and no two patients with a shared diagnosis exhibit identical symptom profiles. Over the last several decades, group-level analyses of in vivo neuroimaging data have led to fundamental advances in our understanding of the neurobiology of psychiatric illnesses. More recently, access to computational resources and large, publicly available datasets alongside the rise of predictive modeling and precision medicine approaches have facilitated the study of psychiatric illnesses at an individual level. Data-driven machine learning analyses can be applied to identify disease-relevant biological subtypes, predict individual symptom profiles, and recommend personalized therapeutic interventions. However, when developing these predictive models, methodological choices must be carefully considered to ensure accurate, robust, and interpretable results. Choices pertaining to algorithms, neuroimaging modalities and states, data transformation, phenotypes, parcellations, sample sizes, and populations we are specifically studying can influence model performance. Here, we review applications of neuroimaging-based machine learning models to study psychiatric illnesses and discuss the effects of different methodological choices on model performance. An understanding of these effects is crucial for the proper implementation of predictive models in psychiatry and will facilitate more accurate diagnoses, prognoses, and therapeutics.

Functional magnetic resonance imaging in migraine: A systematic review

BACKGROUND

Migraine is a highly prevalent primary headache disorder. Despite a high burden of disease, key disease mechanisms are not entirely understood. Functional magnetic resonance imaging is an imaging method using the blood-oxygen-level-dependent signal, which has been increasingly used in migraine research over recent years. This systematic review summarizes recent findings employing functional magnetic resonance imaging for the investigation of migraine.

METHODS

We conducted a systematic search and selection of functional magnetic resonance imaging applications in migraine from April 2014 to December 2021 (PubMed and references of identified articles according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines). Methodological details and main findings were extracted and synthesized.

RESULTS

Out of 224 articles identified, 114 were included after selection. Repeatedly emerging structures of interest included the insula, brainstem, limbic system, hypothalamus, thalamus, and functional networks. Assessment of functional brain changes in response to treatment is emerging, and machine learning has been used to investigate potential functional magnetic resonance imaging-based markers of migraine.

CONCLUSIONS

A wide variety of functional magnetic resonance imaging-based metrics were found altered across the brain for heterogeneous migraine cohorts, partially correlating with clinical parameters and supporting the concept to conceive migraine as a brain state. However, a majority of findings from previous studies have not been replicated, and studies varied considerably regarding image acquisition and analyses techniques. Thus, while functional magnetic resonance imaging appears to have the potential to advance our understanding of migraine pathophysiology, replication of findings in large representative datasets and precise, standardized reporting of clinical data would likely benefit the field and further increase the value of observations.

Applying joint graph embedding to study Alzheimer's neurodegeneration patterns in volumetric data

Neurodegeneration measured through volumetry in MRI is recognized as a potential Alzheimer's Disease (AD) biomarker, but its utility is limited by lack of specificity. Quantifying spatial patterns of neurodegeneration on a whole brain scale rather than locally may help address this. In this work, we turn to network based analyses and extend a graph embedding algorithm to study morphometric connectivity from volume-change correlations measured with structural MRI on the timescale of years. We model our data with the multiple random eigengraphs framework, as well as modify and implement a multigraph embedding algorithm proposed earlier to estimate a low dimensional embedding of the networks. Our version of the algorithm guarantees meaningful finite-sample results and estimates maximum likelihood edge probabilities from population-specific network modes and subject-specific loadings. Furthermore, we propose and implement a novel statistical testing procedure to analyze group differences after accounting for confounders and locate significant structures during AD neurodegeneration. Family-wise error rate is controlled at 5% using permutation testing on the maximum statistic. We show that results from our analysis reveal networks dominated by known structures associated to AD neurodegeneration, indicating the framework has promise for studying AD. Furthermore, we find network-structure tuples that are not found with traditional methods in the field.

The classification of brain network for major depressive disorder patients based on deep graph convolutional neural network

Introduction

The early diagnosis of major depressive disorder (MDD) is very important for patients that suffer from severe and irreversible consequences of depression. It has been indicated that functional connectivity (FC) analysis based on functional magnetic resonance imaging (fMRI) data can provide valuable biomarkers for clinical diagnosis. However, previous studies mainly focus on brain disease classification in small sample sizes, which may lead to dramatic divergences in classification accuracy.

Methods

This paper attempts to address this limitation by applying the deep graph convolutional neural network (DGCNN) method on a large multi-site MDD dataset. The resting-state fMRI data are acquired from 830 MDD patients and 771 normal controls (NC) shared by the REST-meta-MDD consortium.

Results

The DGCNN model trained with the binary network after thresholding, identified MDD patients from normal controls and achieved an accuracy of 72.1% with 10-fold cross-validation, which is 12.4%, 9.8%, and 7.6% higher than SVM, RF, and GCN, respectively. Moreover, the process of dataset reading and model training is faster. Therefore, it demonstrates the advantages of the DGCNN model with low time complexity and sound classification performance.

Discussion

Based on a large, multi-site dataset from MDD patients, the results expressed that DGCNN is not an extremely accurate method for MDD diagnosis. However, there is an improvement over previous methods with our goal of better understanding brain function and ultimately providing a biomarker or diagnostic capability for MDD diagnosis.

Resting state network mapping in individuals using deep learning

Introduction

Resting state functional MRI (RS-fMRI) is currently used in numerous clinical and research settings. The localization of resting state networks (RSNs) has been utilized in applications ranging from group analysis of neurodegenerative diseases to individual network mapping for pre-surgical planning of tumor resections. Reproducibility of these results has been shown to require a substantial amount of high-quality data, which is not often available in clinical or research settings.

Methods

In this work, we report voxelwise mapping of a standard set of RSNs using a novel deep 3D convolutional neural network (3DCNN). The 3DCNN was trained on publicly available functional MRI data acquired in n = 2010 healthy participants. After training, maps that represent the probability of a voxel belonging to a particular RSN were generated for each participant, and then used to calculate mean and standard deviation (STD) probability maps, which are made publicly available. Further, we compared our results to previously published resting state and task-based functional mappings.

Results

Our results indicate this method can be applied in individual subjects and is highly resistant to both noisy data and fewer RS-fMRI time points than are typically acquired. Further, our results show core regions within each network that exhibit high average probability and low STD.

Discussion

The 3DCNN algorithm can generate individual RSN localization maps, which are necessary for clinical applications. The similarity between 3DCNN mapping results and task-based fMRI responses supports the association of specific functional tasks with RSNs.

Pre- and post-task resting-state differs in clinical populations

Resting-state functional connectivity has generated great hopes as a potential brain biomarker for improving prevention, diagnosis, and treatment in psychiatry. This neuroimaging protocol can routinely be performed by patients and does not depend on the specificities of a task. Thus, it seems ideal for big data approaches that require aggregating data across multiple studies and sites. However, technical variability, diverging data analysis approaches, and differences in data acquisition protocols introduce heterogeneity to the aggregated data. Besides these technical aspects, a prior task that changes the psychological state of participants might also contribute to heterogeneity. In healthy participants, studies have shown that behavioral tasks can influence resting-state measures, but such effects have not yet been reported in clinical populations. Here, we fill this knowledge gap by comparing resting-state functional connectivity before and after clinically relevant tasks in two clinical conditions, namely substance use disorders and phobias. The tasks consisted of viewing craving-inducing and spider anxiety provoking pictures that are frequently used in cue-reactivity studies and exposure therapy. We found distinct pre- vs post-task resting-state connectivity differences in each group, as well as decreased thalamo-cortical and increased intra-thalamic connectivity which might be associated with decreased vigilance in both groups. Our results confirm that resting-state measures can be strongly influenced by prior emotion-inducing tasks that need to be taken into account when pooling resting-state scans for clinical biomarker detection. This demands that resting-state datasets should include a complete description of the experimental design, especially when a task preceded data collection.

Regional brain dysfunction in insomnia after ischemic stroke: A resting-state fMRI study

Objective

This study aimed to explore the abnormality of local brain function in patients with post-stroke insomnia (PSI) based on fMRI and explore the possible neuropathological mechanisms of insomnia in patients with PSI in combination with the Pittsburgh sleep quality index (PSQI) score and provide an objective evaluation index for the follow-up study of acupuncture treatment of PSI.

Methods

A total of 27 patients with insomnia after stroke were enrolled, and the PSQI was used to evaluate their sleep status. Twenty-seven healthy participants who underwent physical examinations during the same period were selected as controls. Resting-state brain function images and structural images of the two groups of participants were collected, and the abnormal changes in the regional brain function in patients with PSI were analyzed using three methods: regional homogeneity (ReHo), the amplitude of low-frequency fluctuations (ALFF) and fractional ALFF (fALFF), and a correlation analysis with the PSQI scale score.

Results

Compared with the HCs, the ReHo values of the PSI group in the bilateral lingual gyrus, right cuneus, right precentral and postcentral gyri were significantly lower, and the ReHo values of the left supramarginal gyrus were significantly higher. In the PSI group, the ALFF values in the bilateral lingual gyrus were significantly decreased, whereas those in the bilateral middle temporal gyrus, right inferior temporal gyrus, right inferior frontal gyrus, right limbic lobe, right precuneus, left posterior cingulate gyrus, and left middle occipital gyrus were significantly increased. Compared with HCs, the fALFF values of the bilateral lingual gyrus, bilateral inferior occipital gyrus, and bilateral cuneus in the PSI group were significantly higher. The ReHo value of the left supramarginal gyrus in the PSI group was significantly negatively correlated with the total PSQI score.

Conclusion

Patients with PSI have abnormal local activities in multiple brain regions, including the visual processing-related cortex, sensorimotor cortex, and some default-mode network (DMN) regions. Over-arousal of the DMN and over-sensitivity of the audiovisual stimuli in patients with PSI may be the main mechanisms of insomnia and can lead to a decline in cognitive function and abnormalities in emotion regulation simultaneously.

Identification of Resting-State Network Functional Connectivity and Brain Structural Signatures in Fibromyalgia Using a Machine Learning Approach

Abnormal resting-state functional connectivity (rs-FC) and brain structure have emerged as pathological hallmarks of fibromyalgia (FM). This study investigated and compared the accuracy of network rs-FC and brain structural features in identifying FM with a machine learning (ML) approach. Twenty-six FM patients and thirty healthy controls were recruited. Clinical presentation was measured by questionnaires. After MRI acquisitions, network rs-FC z-score and network-based gray matter volume matrices were exacted and preprocessed. The performance of feature selection and classification methods was measured. Correlation analyses between predictive features in final models and clinical data were performed. The combination of the recursive feature elimination (RFE) selection method and support vector machine (rs-FC data) or logistic regression (structural data), after permutation importance feature selection, showed high performance in distinguishing FM patients from pain-free controls, in which the rs-FC ML model outperformed the structural ML model (accuracy: 0.91 vs. 0.86, AUC: 0.93 vs. 0.88). The combined rs-FC and structural ML model showed the best performance (accuracy: 0.95, AUC: 0.95). Additionally, several rs-FC features in the final ML model correlated with FM's clinical data. In conclusion, ML models based on rs-FC and brain structural MRI features could effectively differentiate FM patients from pain-free subjects.

Superior temporal gyrus functional connectivity predicts transcranial direct current stimulation response in Schizophrenia: A machine learning study

Transcranial direct current stimulation (tDCS) is a promising adjuvant treatment for persistent auditory verbal hallucinations (AVH) in Schizophrenia (SZ). Nonetheless, there is considerable inter-patient variability in the treatment response of AVH to tDCS in SZ. Machine-learned models have the potential to predict clinical response to tDCS in SZ. This study aims to examine the feasibility of identifying SZ patients with persistent AVH (SZ-AVH) who will respond to tDCS based on resting-state functional connectivity (rs-FC). Thirty-four SZ-AVH patients underwent resting-state functional MRI at baseline followed by add-on, twice-daily, 20-min sessions with tDCS (conventional/high-definition) for 5 days. A machine learning model was developed to identify tDCS treatment responders based on the rs-FC pattern, using the left superior temporal gyrus (LSTG) as the seed region. Functional connectivity between LSTG and brain regions involved in auditory and sensorimotor processing emerged as the important predictors of the tDCS treatment response. L1-regularized logistic regression model had an overall accuracy of 72.5% in classifying responders vs. non-responders. This model outperformed the state-of-the-art convolutional neural networks (CNN) model-both without (59.41%) and with pre-training (68.82%). It also outperformed the L1-logistic regression model trained with baseline demographic features and clinical scores of SZ patients. This study reports the first evidence that rs-fMRI-derived brain connectivity pattern can predict the clinical response of persistent AVH to add-on tDCS in SZ patients with 72.5% accuracy.

Predicting response to tVNS in patients with migraine using functional MRI: A voxels-based machine learning analysis

Background

Migraine is a common disorder, affecting many patients. However, for one thing, lacking objective biomarkers, misdiagnosis, and missed diagnosis happen occasionally. For another, though transcutaneous vagus nerve stimulation (tVNS) could alleviate migraine symptoms, the individual difference of tVNS efficacy in migraineurs hamper the clinical application of tVNS. Therefore, it is necessary to identify biomarkers to discriminate migraineurs as well as select patients suitable for tVNS treatment.

Methods

A total of 70 patients diagnosed with migraine without aura (MWoA) and 70 matched healthy controls were recruited to complete fMRI scanning. In study 1, the fractional amplitude of low-frequency fluctuation (fALFF) of each voxel was calculated, and the differences between healthy controls and MWoA were compared. Meaningful voxels were extracted as features for discriminating model construction by a support vector machine. The performance of the discriminating model was assessed by accuracy, sensitivity, and specificity. In addition, a mask of these significant brain regions was generated for further analysis. Then, in study 2, 33 of the 70 patients with MWoA in study 1 receiving real tVNS were included to construct the predicting model in the generated mask. Discriminative features of the discriminating model in study 1 were used to predict the reduction of attack frequency after a 4-week tVNS treatment by support vector regression. A correlation coefficient between predicted value and actual value of the reduction of migraine attack frequency was conducted in 33 patients to assess the performance of predicting model after tVNS treatment. We vislized the distribution of the predictive voxels as well as investigated the association between fALFF change (post-per treatment) of predict weight brain regions and clinical outcomes (frequency of migraine attack) in the real group.

Results

A biomarker containing 3,650 features was identified with an accuracy of 79.3%, sensitivity of 78.6%, and specificity of 80.0% (p &lt; 0.002). The discriminative features were found in the trigeminal cervical complex/rostral ventromedial medulla (TCC/RVM), thalamus, medial prefrontal cortex (mPFC), and temporal gyrus. Then, 70 of 3,650 discriminative features were identified to predict the reduction of attack frequency after tVNS treatment with a correlation coefficient of 0.36 (p = 0.03). The 70 predictive features were involved in TCC/RVM, mPFC, temporal gyrus, middle cingulate cortex (MCC), and insula. The reduction of migraine attack frequency had a positive correlation with right TCC/RVM (r = 0.433, p = 0.021), left MCC (r = 0.451, p = 0.016), and bilateral mPFC (r = 0.416, p = 0.028), and negative with left insula (r = −0.473, p = 0.011) and right superior temporal gyrus/middle temporal gyrus (r = −0.684, p &lt; 0.001), respectively.

Conclusions

By machine learning, the study proposed two potential biomarkers that could discriminate patients with MWoA and predict the efficacy of tVNS in reducing migraine attack frequency. The pivotal features were mainly located in the TCC/RVM, thalamus, mPFC, and temporal gyrus.

Influence of Gestational Diabetes and Pregestational Maternal BMI on the Brain of Six-Year-Old Offspring

BACKGROUND

Gestational diabetes (GD) and maternal excess weight are common pregnancy conditions that increase the risk of future complications for both the mother and her offspring. Their consequences on neurodevelopment are widely described in the literature, but less is known concerning the potential transgenerational influence on the brain structure.

METHODS

We used a combination of support vectors machine and hierarchical clustering to investigate the potential presence of anatomical brain differences in a sample of 109 children aged six years, born to mothers with overweight or obesity, or to mothers diagnosed with GD during pregnancy.

RESULTS

Significant effects are visible in the brain of children born to mothers with GD associated with pregestational excess weight, especially overweight instead of obesity. No differences in children's brain were observed when considering those born to normal-weight mothers.

CONCLUSIONS

Our study highlights the need for clinical attention of pregnant women at risk to develop GD, and especially those with pregestational excess weight, since this status was found to be associated with detectable transgenerational brain changes. These effects may be due to the absence of specific and individualized intervention in these mothers during pregnancy.

A Deep Spatiotemporal Attention Network for Mild Cognitive Impairment Identification

Mild cognitive impairment (MCI) is a nervous system disease, and its clinical status can be used as an early warning of Alzheimer's disease (AD). Subtle and slow changes in brain structure between patients with MCI and normal controls (NCs) deprive them of effective diagnostic methods. Therefore, the identification of MCI is a challenging task. The current functional brain network (FBN) analysis to predict human brain tissue structure is a new method emerging in recent years, which provides sensitive and effective medical biomarkers for the diagnosis of neurological diseases. Therefore, to address this challenge, we propose a novel Deep Spatiotemporal Attention Network (DSTAN) framework for MCI recognition based on brain functional networks. Specifically, we first extract spatiotemporal features between brain functional signals and FBNs by designing a spatiotemporal convolution strategy (ST-CONV). Then, on this basis, we introduce a learned attention mechanism to further capture brain nodes strongly correlated with MCI. Finally, we fuse spatiotemporal features for MCI recognition. The entire network is trained in an end-to-end fashion. Extensive experiments show that our proposed method significantly outperforms current baselines and state-of-the-art methods, with a classification accuracy of 84.21%.

Somatosensory Deficits After Stroke: Insights From MRI Studies

Somatosensory deficits after stroke are a major health problem, which can impair patients' health status and quality of life. With the developments in human brain mapping techniques, particularly magnetic resonance imaging (MRI), many studies have applied those techniques to unravel neural substrates linked to apoplexy sequelae. Multi-parametric MRI is a vital method for the measurement of stroke and has been applied to diagnose stroke severity, predict outcome and visualize changes in activation patterns during stroke recovery. However, relatively little is known about the somatosensory deficits after stroke and their recovery. This review aims to highlight the utility and importance of MRI techniques in the field of somatosensory deficits and synthesizes corresponding articles to elucidate the mechanisms underlying the occurrence and recovery of somatosensory symptoms. Here, we start by reviewing the anatomic and functional features of the somatosensory system. And then, we provide a discussion of MRI techniques and analysis methods. Meanwhile, we present the application of those techniques and methods in clinical studies, focusing on recent research advances and the potential for clinical translation. Finally, we identify some limitations and open questions of current imaging studies that need to be addressed in future research.

A semi-supervised classification RBM with an improved fMRI representation algorithm

BACKGROUND AND OBJECTIVE

Training an effective and robust supervised learning classifier is not easy due to the limitations of acquiring and labeling considerable human functional magnetic resonance imaging (fMRI) data. Semi-supervised learning uses unlabeled data for feature learning and combines them into labeled data to build better classification models.

METHODS

Since no premises or assumptions are required, a restricted Boltzmann machine (RBM) is suitable for learning data representation of neuroimages. In our study, an improved fMRI representation algorithm with a hybrid L1/L2 regularization method (HRBM) was proposed to optimize the original model for sparsity. Different from common semi-supervised classification models that treat feature learning and classification as two separate training steps, we then constructed a new semi-supervised classification RBM based on a joint training algorithm with HRBM, named Semi-HRBM. This joint training algorithm jointly trains the objective function of feature learning and classification process, so that the learned features can effectively represent the original fMRI data and adapt to the classification tasks.

RESULTS

This study uses fMRI data to identify categories of visual stimuli. In the fMRI data classification task under four visual stimuli (house, face, car, and cat), our HRBM has satisfactory feature representation capabilities and better performance for multiple classification tasks. Taking the supervised RBM (sup-RBM) as an example, our Semi-HRBM classification model improves the average accuracy of the four-classification task by 7.68%, and improves the average F1 score of each visual stimulus task by 8.90%. In addition, the generalization ability of the model was also improved.

CONCLUSION

This research might contribute to enrich solutions for insufficiently labeled neuroimaging samples, which could help to identify complex brain states under different stimuli or tasks.

Behavioral Studies Using Large-Scale Brain Networks – Methods and Validations

Mapping human behaviors to brain activity has become a key focus in modern cognitive neuroscience. As methods such as functional MRI (fMRI) advance cognitive scientists show an increasing interest in investigating neural activity in terms of functional connectivity and brain networks, rather than activation in a single brain region. Due to the noisy nature of neural activity, determining how behaviors are associated with specific neural signals is not well-established. Previous research has suggested graph theory techniques as a solution. Graph theory provides an opportunity to interpret human behaviors in terms of the topological organization of brain network architecture. Graph theory-based approaches, however, only scratch the surface of what neural connections relate to human behavior. Recently, the development of data-driven methods, e.g., machine learning and deep learning approaches, provide a new perspective to study the relationship between brain networks and human behaviors across the whole brain, expanding upon past literatures. In this review, we sought to revisit these data-driven approaches to facilitate our understanding of neural mechanisms and build models of human behaviors. We start with the popular graph theory approach and then discuss other data-driven approaches such as connectome-based predictive modeling, multivariate pattern analysis, network dynamic modeling, and deep learning techniques that quantify meaningful networks and connectivity related to cognition and behaviors. Importantly, for each topic, we discuss the pros and cons of the methods in addition to providing examples using our own data for each technique to describe how these methods can be applied to real-world neuroimaging data.

Cerebral Small Vessel Disease: Neuroimaging Features, Biochemical Markers, Influencing Factors, Pathological Mechanism and Treatment

Cerebral small vessel disease (CSVD) is the most common chronic vascular disease involving the whole brain. Great progress has been made in clinical imaging, pathological mechanism, and treatment of CSVD, but many problems remain. Clarifying the current research dilemmas and future development direction of CSVD can provide new ideas for both basic and clinical research. In this review, the risk factors, biological markers, pathological mechanisms, and the treatment of CSVD will be systematically illustrated to provide the current research status of CSVD. The future development direction of CSVD will be elucidated by summarizing the research difficulties.

Collaborative learning of graph generation, clustering and classification for brain networks diagnosis

PURPOSE

Accurate diagnosis of autism spectrum disorder (ASD) plays a key role in improving the condition and quality of life for patients. In this study, we mainly focus on ASD diagnosis with functional brain networks (FBNs). The major challenge for brain networks modeling is the high dimensional connectivity in brain networks and limited number of subjects, which hinders the classification capability of graph convolutional networks (GCNs).

METHOD

To alleviate the influence of the limited data and high dimensional connectivity, we introduce a unified three-stage graph learning framework for brain network classification, involving multi-graph clustering, graph generation and graph classification. The framework combining Graph Generation, Clustering and Classification Networks (GraphCGC-Net) enhances the critical connections by multi-graph clustering (MGC) with a supervision scheme, and generates realistic brain networks by simultaneously preserving the global consistent distribution and local topology properties.

RESULTS

To demonstrate the effectiveness of our approach, we evaluate the performance of the proposed method on the Autism Brain Imaging Data Exchange (ABIDE) dataset and conduct extensive experiments on the ASD classification problem. Our proposed method achieves an average accuracy of 70.45% and an AUC of 72.76% on ABIDE. Compared with the traditional GCN model, the proposed GraphCGC-Net obtains 9.3%, and 10.64% improvement in terms of accuracy and AUC metrics, respectively.

CONCLUSION

The comprehensive experiments demonstrate that our GraphCGC-Net is effective for graph classification in brain disorders diagnosis. Moreover, we find that MGC can generate biologically meaningful subnetworks, which is highly consistent with the previous neuroimaging-derived biomarker evidence of ASD. More importantly, the promising results suggest that applying generative adversarial networks (GANs) in brain networks to improve the classification performance is worth further investigation.

Abnormal Fractional Amplitude of Low Frequency Fluctuation Changes in Patients With Dry Eye Disease: A Functional Magnetic Resonance Imaging Study

Purpose

To investigate spontaneous brain activity in patients with dry eye (DE) and healthy control (HC) using the fractional amplitude of low frequency fluctuation (fALFF) technique with the aim of elucidating the relationship between the clinical symptoms of DE and changes in brain function.

Material and Methods

A total of 28 patients with DE and 28 matched healthy volunteers (10 males and 18 females in each group) were enrolled. Resting-state functional magnetic resonance imaging scans were performed in both groups. Then all subjects were required to complete a comprehensive Hospital Anxiety and Depression Scale (HADS). Receiver operating characteristic (ROC) curve analysis was used to evaluate the differences in fALFF values between the two groups and their diagnostic value. Linear correlations between HADS and fALFF values in different brain regions of DE patients were analyzed using the Pearson correlation coefficient.

Results

Patients with DE had significantly higher fALFF values in the left calcarine sulcus (CS) than the HC group, while fALFF values in the bilateral middle frontal gyrus (MFG) and right MFG/right inferior frontal gyrus (IFG) were significantly lower in DE patients than in HC group. fALFF values had a high diagnostic value for differentiating patients with DE from the HC group (P &lt; 0.001). Right MFG and right MFG/IFG were significantly correlated with HADS values.

Conclusion

Our study found that DE mainly involved functional disorders in the brain areas of the left CS, bilateral MFG and right MFG/right IFG, which helped us to find possible clinical features of DE disease and reflected the potential pathological mechanism of DE.

Biomarkers identification for Schizophrenia via VAE and GSDAE-based data augmentation

Deep learning has made great progress in analyzing MRI data, while the MRI data with high dimensional but small sample size (HDSSS) brings many limitations to biomarkers identification. Few-shot learning has been proposed to solve such problems and data augmentation is a typical method of it. The variational auto-encoder (VAE) is a generative method based on variational Bayesian inference that is used for data augmentation. Graph regularized sparse deep autoencoder (GSDAE) can reconstruct sparse samples and keep the manifold structure of data which will facilitate biomarkers selection greatly. To generate better HDSSS data for biomarkers identification, a data augmentation method based on VAE and GSDAE is proposed in this paper, termed GS-VDAE. Instead of utilizing the final products of GSDAE, our proposed model embeds the generation procedure into GSDAE for augmentation. In this way, the augmented samples will be rooted in the significant features extracted from the original samples, which can ensure the newly formed samples contain the most significant characteristics of the original samples. The classification accuracy of the samples generated directly from VAE is 0.74, while the classification accuracy of the samples generated from GS-VDAE is 0.84, which proves the validity of our model. Additionally, a regression feature selection method with truncated nuclear norm regularization is chosen for biomarkers selection. The biomarkers selection results of schizophrenia data reveal that the augmented samples obtained by our proposed method can get higher classification accuracy with less ranked features compared with original samples, which proves the validation of our model.

Multi-level Functional Connectivity Fusion Classification Framework for Brain Disease Diagnosis

Brain disease diagnosis is a new hotspot in the cross research of artificial intelligence and neuroscience. Quantitative analysis of functional magnetic resonance imaging (fMRI) data can provide valuable biomarkers that contributes to clinical diagnosis, and the analysis of functional connectivity (FC) has become the primary method. However, previous studies mainly focus on brain disease classification based on the low-order FC features, ignoring the potential role of high-order functional relationships among brain regions. To solve this problem, this study proposed a novel multi-level FC fusion classification framework (MFC) for brain disease diagnosis. We firstly designed a deep neural network (DNN) model to extract and learn abstract feature representations for the constructed low-order and high-order FC patterns. Both unsupervised and supervised learning steps were performed during the DNN model training, and the prototype learning was introduced in the supervised fine-tuning to improve the intra-class compactness and inter-class separability of the feature representation. Then, we combined the learned multi-level abstract FC features and trained an ensemble classifier with a hierarchical stacking learning strategy for the brain disease classification. Systematic experiments were conducted on two real large-scale fMRI datasets. Results showed that the proposed MFC model obtained robust classification performance for different preprocessing pipelines, different brain parcellations, and different cross-validation schemes, suggesting the effectiveness and generality of the proposed MFC model. Overall, this study provides a promising solution to combine the informative low-order and high-order FC patterns to further promote the classification of brain diseases.