fMRI Datasets Research Articles

Atypical hemispheric lateralization of brain function and structure in autism: a comprehensive meta-analysis study.

Characteristic changes in the asymmetric nature of the human brain are associated with neurodevelopmental differences related to autism. In people with autism, these differences are thought to affect brain structure and function, although the structural and functional bases of these defects are yet to be fully characterized. We applied a comprehensive meta-analysis to resting-state functional and structural magnetic resonance imaging datasets from 370 people with autism and 498 non-autistic controls using seven datasets of the Autism Brain Imaging Data Exchange Project. We studied the meta-effect sizes based on standardized mean differences and standard deviations (s.d.) for lateralization of gray matter volume (GMV), fractional amplitude of low-frequency fluctuation (fALFF), and regional homogeneity (ReHo). We examined the functional correlates of atypical laterality through an indirect annotation approach followed by a direct correlation analysis with symptom scores. In people with autism, 85, 51, and 51% of brain regions showed a significant diagnostic effect for lateralization in GMV, fALFF, and ReHo, respectively. Among these regions, 35.7% showed overlapping differences in lateralization in GMV, fALFF, and ReHo, particularly in regions with functional annotations for language, motor, and perceptual functions. These differences were associated with clinical measures of reciprocal social interaction, communication, and repetitive behaviors. A meta-analysis based on s.d. showed that people with autism had lower variability in structural lateralization but higher variability in functional lateralization. These findings highlight that atypical hemispheric lateralization is a consistent feature in autism across different sites and may be used as a neurobiological marker for autism.

GATE: Graph CCA for Temporal Self-Supervised Learning for Label-Efficient fMRI Analysis.

In this work, we focus on the challenging task, neuro-disease classification, using functional magnetic resonance imaging (fMRI). In population graph-based disease analysis, graph convolutional neural networks (GCNs) have achieved remarkable success. However, these achievements are inseparable from abundant labeled data and sensitive to spurious signals. To improve fMRI representation learning and classification under a label-efficient setting, we propose a novel and theory-driven self-supervised learning (SSL) framework on GCNs, namely Graph CCA for Temporal sElf-supervised learning on fMRI analysis (GATE). Concretely, it is demanding to design a suitable and effective SSL strategy to extract formation and robust features for fMRI. To this end, we investigate several new graph augmentation strategies from fMRI dynamic functional connectives (FC) for SSL training. Further, we leverage canonical-correlation analysis (CCA) on different temporal embeddings and present the theoretical implications. Consequently, this yields a novel two-step GCN learning procedure comprised of (i) SSL on an unlabeled fMRI population graph and (ii) fine-tuning on a small labeled fMRI dataset for a classification task. Our method is tested on two independent fMRI datasets, demonstrating superior performance on autism and dementia diagnosis. Our code is available at https://github.com/LarryUESTC/GATE.

A deep residual model for characterization of 5D spatiotemporal network dynamics reveals widespread spatiodynamic changes in schizophrenia.

Schizophrenia is a severe brain disorder with serious symptoms including delusions, disorganized speech, and hallucinations that can have a long-term detrimental impact on different aspects of a patient's life. It is still unclear what the main cause of schizophrenia is, but a combination of altered brain connectivity and structure may play a role. Neuroimaging data has been useful in characterizing schizophrenia, but there has been very little work focused on voxel-wise changes in multiple brain networks over time, despite evidence that functional networks exhibit complex spatiotemporal changes over time within individual subjects. Recent studies have primarily focused on static (average) features of functional data or on temporal variations between fixed networks; however, such approaches are not able to capture multiple overlapping networks which change at the voxel level. In this work, we employ a deep residual convolutional neural network (CNN) model to extract 53 different spatiotemporal networks each of which captures dynamism within various domains including subcortical, cerebellar, visual, sensori-motor, auditory, cognitive control, and default mode. We apply this approach to study spatiotemporal brain dynamism at the voxel level within multiple functional networks extracted from a large functional magnetic resonance imaging (fMRI) dataset of individuals with schizophrenia (N = 708) and controls (N = 510). Our analysis reveals widespread group level differences across multiple networks and spatiotemporal features including voxel-wise variability, magnitude, and temporal functional network connectivity in widespread regions expected to be impacted by the disorder. We compare with typical average spatial amplitude and show highly structured and neuroanatomically relevant results are missed if one does not consider the voxel-wise spatial dynamics. Importantly, our approach can summarize static, temporal dynamic, spatial dynamic, and spatiotemporal dynamics features, thus proving a powerful approach to unify and compare these various perspectives. In sum, we show the proposed approach highlights the importance of accounting for both temporal and spatial dynamism in whole brain neuroimaging data generally, shows a high-level of sensitivity to schizophrenia highlighting global but spatially unique dynamics showing group differences, and may be especially important in studies focused on the development of brain-based biomarkers.

Quality control practices in FMRI analysis: Philosophy, methods and examples using AFNI.

Quality control (QC) is a necessary, but often an under-appreciated, part of FMRI processing. Here we describe procedures for performing QC on acquired or publicly available FMRI datasets using the widely used AFNI software package. This work is part of the Research Topic, "Demonstrating Quality Control (QC) Procedures in fMRI." We used a sequential, hierarchical approach that contained the following major stages: (1) GTKYD (getting to know your data, esp. its basic acquisition properties), (2) APQUANT (examining quantifiable measures, with thresholds), (3) APQUAL (viewing qualitative images, graphs, and other information in systematic HTML reports) and (4) GUI (checking features interactively with a graphical user interface); and for task data, and (5) STIM (checking stimulus event timing statistics). We describe how these are complementary and reinforce each other to help researchers stay close to their data. We processed and evaluated the provided, publicly available resting state data collections (7 groups, 139 total subjects) and task-based data collection (1 group, 30 subjects). As specified within the Topic guidelines, each subject's dataset was placed into one of three categories: Include, exclude or uncertain. The main focus of this paper, however, is the detailed description of QC procedures: How to understand the contents of an FMRI dataset, to check its contents for appropriateness, to verify processing steps, and to examine potential quality issues. Scripts for the processing and analysis are freely available.

Multiscale functional connectivity patterns of the aging brain learned from harmonized rsfMRI data of the multi-cohort iSTAGING study

To learn multiscale functional connectivity patterns of the aging brain, we built a brain age prediction model of functional connectivity measures at seven scales on a large fMRI dataset, consisting of resting-state fMRI scans of 4186 individuals with a wide age range (22 to 97 years, with an average of 63) from five cohorts. We computed multiscale functional connectivity measures of individual subjects using a personalized functional network computational method, harmonized the functional connectivity measures of subjects from multiple datasets in order to build a functional brain age model, and finally evaluated how functional brain age gap correlated with cognitive measures of individual subjects. Our study has revealed that functional connectivity measures at multiple scales were more informative than those at any single scale for the brain age prediction, the data harmonization significantly improved the brain age prediction performance, and the data harmonization in the functional connectivity measures' tangent space worked better than in their original space. Moreover, brain age gap scores of individual subjects derived from the brain age prediction model were significantly correlated with clinical and cognitive measures. Overall, these results demonstrated that multiscale functional connectivity patterns learned from a large-scale multi-site rsfMRI dataset were informative for characterizing the aging brain and the derived brain age gap was associated with cognitive and clinical measures.

Interaction between Functional Connectivity and Neural Excitability in Autism: A Novel Framework for Computational Modeling and Application to Biological Data.

Functional connectivity (FC) and neural excitability may interact to affect symptoms of autism spectrum disorder (ASD). We tested this hypothesis with neural network simulations, and applied it with functional magnetic resonance imaging (fMRI). A hierarchical recurrent neural network embodying predictive processing theory was subjected to a facial emotion recognition task. Neural network simulations examined the effects of FC and neural excitability on changes in neural representations by developmental learning, and eventually on ASD-like performance. Next, by mapping each neural network condition to subject subgroups on the basis of fMRI parameters, the association between ASD-like performance in the simulation and ASD diagnosis in the corresponding subject subgroup was examined. In the neural network simulation, the more homogeneous the neural excitability of the lower-level network, the more ASD-like the performance (reduced generalization and emotion recognition capability). In addition, in homogeneous networks, the higher the FC, the more ASD-like performance, while in heterogeneous networks, the higher the FC, the less ASD-like performance, demonstrating that FC and neural excitability interact. As an underlying mechanism, neural excitability determines the generalization capability of top-down prediction, and FC determines whether the model's information processing will be top-down prediction-dependent or bottom-up sensory-input dependent. In fMRI datasets, ASD was actually more prevalent in subject subgroups corresponding to the network condition showing ASD-like performance. The current study suggests an interaction between FC and neural excitability, and presents a novel framework for computational modeling and biological application of a developmental learning process underlying cognitive alterations in ASD.

Increased functional connectivity patterns in mild Alzheimer's disease: A rsfMRI study.

Alzheimer's disease (AD) is the most common age-related neurodegenerative disorder. In view of our rapidly aging population, there is an urgent need to identify Alzheimer's disease (AD) at an early stage. A potential way to do so is by assessing the functional connectivity (FC), i.e., the statistical dependency between two or more brain regions, through novel analysis techniques. In the present study, we assessed the static and dynamic FC using different approaches. A resting state (rs)fMRI dataset from the Alzheimer's disease neuroimaging initiative (ADNI) was used (n = 128). The blood-oxygen-level-dependent (BOLD) signals from 116 regions of 4 groups of participants, i.e., healthy controls (HC; n = 35), early mild cognitive impairment (EMCI; n = 29), late mild cognitive impairment (LMCI; n = 30), and Alzheimer's disease (AD; n = 34) were extracted and analyzed. FC and dynamic FC were extracted using Pearson's correlation, sliding-windows correlation analysis (SWA), and the point process analysis (PPA). Additionally, graph theory measures to explore network segregation and integration were computed. Our results showed a longer characteristic path length and a decreased degree of EMCI in comparison to the other groups. Additionally, an increased FC in several regions in LMCI and AD in contrast to HC and EMCI was detected. These results suggest a maladaptive short-term mechanism to maintain cognition. The increased pattern of FC in several regions in LMCI and AD is observable in all the analyses; however, the PPA enabled us to reduce the computational demands and offered new specific dynamic FC findings.

Resting-state functional connectivity of the raphe nuclei in major depressive Disorder: A Multi-site study

Accumulating evidence showed that major depressive disorder (MDD) is characterized by a dysfunction of serotonin neurotransmission. Raphe nuclei are the sources of most serotonergic neurons that project throughout the brain. Incorporating measurements of activity within the raphe nuclei into the analysis of connectivity characteristics may contribute to understanding how neurotransmitter synthesized centers are involved in thepathogenesisof MDD. Here, we analyzed the resting-state functional magnetic resonance imaging (RS-fMRI) dataset from 1,148 MDD patients and 1,079 healthy individuals recruited across nine centers. A seed-based analysis with the dorsal raphe and median raphe nuclei was performed to explore the functional connectivity (FC) alterations. Compared to controls, for dorsal raphe, the significantly decreased FC linking with the right precuneus and median cingulate cortex were found; for median raphe, the increased FC linking with right superior cerebellum (lobules V/VI) was found in MDD patients. In further exploratory analyzes, MDD-related connectivity alterations in dorsal and median raphe nuclei in different clinical factors remained highly similar to the main findings, indicating these abnormal connectivities are a disease-related alteration. Our study highlights a functional dysconnection pattern of raphe nuclei in MDD with multi-site big data. These findings help improve our understanding of the pathophysiology of depression and provide evidence of the theoretical foundation for the development of novel pharmacotherapies.

Altered Resting-State functional connectivity in the anterior and posterior hippocampus in Post-traumatic stress disorder: The central role of the anterior hippocampus

BackgroundPost-traumatic stress disorder can be viewed as a memory disorder, with trauma-related flashbacks being a core symptom. Given the central role of the hippocampus in autobiographical memory, surprisingly, there is mixed evidence concerning altered hippocampal functional connectivity in PTSD. We shed light on this discrepancy by considering the distinct roles of the anterior versus posterior hippocampus and examine how this distinction may map onto whole-brain resting-state functional connectivity patterns among those with and without PTSD. MethodsWe first assessed whole-brain between-group differences in the functional connectivity profiles of the anterior and posterior hippocampus within a publicly available data set of resting-state fMRI data from 31 male Vietnam war veterans diagnosed with PTSD (mean age = 67.6 years, sd = 2.3) and 29 age-matched combat-exposed male controls (age = 69.1 years, sd = 3.5). Next, the connectivity patterns of each subject within the PTSD group were correlated with their PTSD symptom scores. Finally, the between-group differences in whole-brain functional connectivity profiles discovered for the anterior and posterior hippocampal seeds were used to prescribe post-hoc ROIs, which were then used to perform ROI-to-ROI functional connectivity and graph-theoretic analyses. ResultsThe PTSD group showed increased functional connectivity of the anterior hippocampus with affective brain regions (anterior/posterior insula, orbitofrontal cortex, temporal pole) and decreased functional connectivity of the anterior/posterior hippocampus with regions involved in processing bodily self-consciousness (supramarginal gyrus). Notably, decreased anterior hippocampus connectivity with the posterior cingulate cortex/precuneus was associated with increased PTSD symptom severity. The left anterior hippocampus also emerged as a central locus of abnormal functional connectivity, with graph-theoretic measures suggestive of a more central hub-like role for this region in those with PTSD compared to trauma-exposed controls. ConclusionsOur results highlight that the anterior hippocampus plays a critical role in the neurocircuitry underlying PTSD and underscore the importance of the differential roles of hippocampal sub-regions in serving as biomarkers of PTSD. Future studies should investigate whether the differential patterns of functional connectivity stemming from hippocampal sub-regions is observed in PTSD populations other than older war veterans.

A transformer model for learning spatiotemporal contextual representation in fMRI data.

Representation learning is a core component in data-driven modeling of various complex phenomena. Learning a contextually informative representation can especially benefit the analysis of fMRI data because of the complexities and dynamic dependencies present in such datasets. In this work, we propose a framework based on transformer models to learn an embedding of the fMRI data by taking the spatiotemporal contextual information in the data into account. This approach takes the multivariate BOLD time series of the regions of the brain as well as their functional connectivity network simultaneously as the input to create a set of meaningful features that can in turn be used in various downstream tasks such as classification, feature extraction, and statistical analysis. The proposed spatiotemporal framework uses the attention mechanism as well as the graph convolution neural network to jointly inject the contextual information regarding the dynamics in time series data and their connectivity into the representation. We demonstrate the benefits of this framework by applying it to two resting-state fMRI datasets, and provide further discussion on various aspects and advantages of it over a number of other commonly adopted architectures.

Contributions of human amygdala nuclei to resting-state networks.

The amygdala is a brain region with a complex internal structure that is associated with psychiatric disease. Methodological limitations have complicated the study of the internal structure of the amygdala in humans. In the current study we examined the functional connectivity between nine amygdaloid nuclei and existing resting-state networks using a high spatial-resolution fMRI dataset. Using data-driven analysis techniques we found that there were three main clusters inside the amygdala that correlated with the somatomotor, ventral attention and default mode networks. In addition, we found that each resting-state networks depended on a specific configuration of amygdaloid nuclei. Finally, we found that co-activity in the cortical-nucleus increased with the severity of self-rated fear in participants. These results highlight the complex nature of amygdaloid connectivity that is not confined to traditional large-scale divisions, implicates specific configurations of nuclei with certain resting-state networks and highlights the potential clinical relevance of the cortical-nucleus in future studies of the human amygdala.

Color-biased regions in the ventral visual pathway are food selective

Color-biased regions have been found between face- and place-selective areas in the ventral visual pathway. To investigate the function of the color-biased regions in a pathway responsible for object recognition, we analyzed the natural scenes dataset (NSD), a large 7T fMRI dataset from 8 participants who each viewed up to 30,000 trials of images of colored natural scenes over more than 30 scanning sessions. In a whole-brain analysis, we correlated the average color saturation of the images with voxel responses, revealing color-biased regions that diverge into two streams, beginning in V4 and extending medially and laterally relative to the fusiform face area in both hemispheres. We drew regions of interest (ROIs) for the two streams and found that the images for each ROI that evoked the largest responses had certain characteristics: they contained food, circular objects, warmer hues, and had higher color saturation. Further analyses showed that food images were the strongest predictor of activity in these regions, implying the existence of medial and lateral ventral food streams (VFSs). We found that color also contributed independently to voxel responses, suggesting that the medial and lateral VFSs use both color and form to represent food. Our findings illustrate how high-resolution datasets such as the NSD can be used to disentangle the multifaceted contributions of many visual features to the neural representations of natural scenes.

Fetal behavior during MRI changes with age and relates to network dynamics.

Fetal motor behavior is an important clinical indicator of healthy development. However, our understanding of associations between fetal behavior and fetal brain development is limited. To fill this gap, this study introduced an approach to automatically and objectively classify long durations of fetal movement from a continuous four-dimensional functional magnetic resonance imaging (fMRI) data set, and paired behavior features with brain activity indicated by the fMRI time series. Twelve-minute fMRI scans were conducted in 120 normal fetuses. Postnatal motor function was evaluated at 7 and 36 months age. Fetal motor behavior was quantified by calculating the frame-wise displacement (FD) of fetal brains extracted by a deep-learning model along the whole time series. Analyzing only low motion data, we characterized the recurring coactivation patterns (CAPs) of the supplementary motor area (SMA). Results showed reduced motor activity with advancing gestational age (GA), likely due in part to loss of space (r=-.51, p < .001). Evaluation of individual variation in motor movement revealed a negative association between movement and the occurrence of coactivations within the left parietotemporal network, controlling for age and sex (p=.003). Further, we found that the occurrence of coactivations between the SMA to posterior brain regions, including visual cortex, was prospectively associated with postnatal motor function at 7months (r=.43, p=.03). This is the first study to pair fetal movement and fMRI, highlighting potential for comparisons of fetal behavior and neural network development to enhance our understanding of fetal brain organization.

Selective and replicable neuroimaging-based indicators of pain discriminability

Neural indicators of pain discriminability have far-reaching theoretical and clinical implications but have been largely overlooked previously. Here, to directly identify the neural basis of pain discriminability, we apply signal detection theory to three EEG (Datasets 1-3, total N= 366) and two fMRI (Datasets 4-5, total N=399) datasets where participants receive transient stimuli of four sensory modalities (pain, touch, audition, and vision) and two intensities (high and low) and report perceptual ratings. Datasets 1 and 4 are used for exploration and others for validation. We find that most pain-evoked EEG and fMRI brain responses robustly encode pain discriminability, which is well replicated in validation datasets. The neural indicators are also pain selective since they cannot track tactile, auditory, or visual discriminability, even though perceptual ratings and sensory discriminability are well matched between modalities. Overall, we provide compelling evidence that pain-evoked brain responses can serve as replicable and selective neural indicators of pain discriminability.

Improving the accuracy of single-trial fMRI response estimates using GLMsingle.

Advances in artificial intelligence have inspired a paradigm shift in human neuroscience, yielding large-scale functional magnetic resonance imaging (fMRI) datasets that provide high-resolution brain responses to thousands of naturalistic visual stimuli. Because such experiments necessarily involve brief stimulus durations and few repetitions of each stimulus, achieving sufficient signal-to-noise ratio can be a major challenge. We address this challenge by introducing GLMsingle, a scalable, user-friendly toolbox available in MATLAB and Python that enables accurate estimation of single-trial fMRI responses (glmsingle.org). Requiring only fMRI time-series data and a design matrix as inputs, GLMsingle integrates three techniques for improving the accuracy of trial-wise general linear model (GLM) beta estimates. First, for each voxel, a custom hemodynamic response function (HRF) is identified from a library of candidate functions. Second, cross-validation is used to derive a set of noise regressors from voxels unrelated to the experiment. Third, to improve the stability of beta estimates for closely spaced trials, betas are regularized on a voxel-wise basis using ridge regression. Applying GLMsingle to the Natural Scenes Dataset and BOLD5000, we find that GLMsingle substantially improves the reliability of beta estimates across visually-responsive cortex in all subjects. Comparable improvements in reliability are also observed in a smaller-scale auditory dataset from the StudyForrest experiment. These improvements translate into tangible benefits for higher-level analyses relevant to systems and cognitive neuroscience. We demonstrate that GLMsingle: (i) helps decorrelate response estimates between trials nearby in time; (ii) enhances representational similarity between subjects within and across datasets; and (iii) boosts one-versus-many decoding of visual stimuli. GLMsingle is a publicly available tool that can significantly improve the quality of past, present, and future neuroimaging datasets sampling brain activity across many experimental conditions.

Federated Analysis in COINSTAC Reveals Functional Network Connectivity and Spectral Links to Smoking and Alcohol Consumption in Nearly 2,000 Adolescent Brains.

With the growth of decentralized/federated analysis approaches in neuroimaging, the opportunities to study brain disorders using data from multiple sites has grown multi-fold. One such initiative is the Neuromark, a fully automated spatially constrained independent component analysis (ICA) that is used to link brain network abnormalities among different datasets, studies, and disorders while leveraging subject-specific networks. In this study, we implement the neuromark pipeline in COINSTAC, an open-source neuroimaging framework for collaborative/decentralized analysis. Decentralized exploratory analysis of nearly 2000 resting-state functional magnetic resonance imaging datasets collected at different sites across two cohorts and co-located in different countries was performed to study the resting brain functional network connectivity changes in adolescents who smoke and consume alcohol. Results showed hypoconnectivity across the majority of networks including sensory, default mode, and subcortical domains, more for alcohol than smoking, and decreased low frequency power. These findings suggest that global reduced synchronization is associated with both tobacco and alcohol use. This proof-of-concept work demonstrates the utility and incentives associated with large-scale decentralized collaborations spanning multiple sites.

Functional Connectome of the Human Brain with Total Correlation.

Recent studies proposed the use of Total Correlation to describe functional connectivity among brain regions as a multivariate alternative to conventional pairwise measures such as correlation or mutual information. In this work, we build on this idea to infer a large-scale (whole-brain) connectivity network based on Total Correlation and show the possibility of using this kind of network as biomarkers of brain alterations. In particular, this work uses Correlation Explanation (CorEx) to estimate Total Correlation. First, we prove that CorEx estimates of Total Correlation and clustering results are trustable compared to ground truth values. Second, the inferred large-scale connectivity network extracted from the more extensive open fMRI datasets is consistent with existing neuroscience studies, but, interestingly, can estimate additional relations beyond pairwise regions. And finally, we show how the connectivity graphs based on Total Correlation can also be an effective tool to aid in the discovery of brain diseases.

An fMRI Dataset for Concept Representation with Semantic Feature Annotations

The neural representation of concepts is a focus of many cognitive neuroscience studies. Prior works studying concept representation with neural imaging data have been largely limited to concrete concepts. The use of relatively small and constrained sets of stimuli leaves open the question of whether the findings can generalize other concepts. We share an fMRI dataset in which 11 participants thought of 672 individual concepts, including both concrete and abstract concepts. The concepts were probed using words paired with images in which the words were selected to cover a wide range of semantic categories. Furthermore, according to the componential theories of concept representation, we collected the 54 semantic features of the 672 concepts comprising sensory, motor, spatial, temporal, affective, social, and cognitive experiences by crowdsourcing annotations. The quality assessment results verify this as a high-quality neuroimaging dataset. Such a dataset is well suited to study how the brain represents different semantic features and concepts, creating the essential condition to investigate the neural representation of individual concepts.

Thalamocortical contributions to working memory processes during the n-back task

Working memory allows individuals to temporally maintain and manipulate information that is no longer accessible from the sensorium. Whereas prior studies have detailed frontoparietal contributions to working memory processes, less emphasis has been placed on subcortical regions, in particular the human thalamus. The thalamus has a complex anatomy that consists of several distinct nuclei, many of which have dense anatomical connectivity with frontoparietal regions, and thus might play an important yet underspecified role for working memory. The goal of our study is to characterize the detailed functional neuroanatomy of the human thalamus and thalamocortical interactions during the n-back task. To that end, we analyzed an n-back fMRI dataset consisting of 395 subjects from the Human Connectome Project (HCP). We found that thalamic nuclei in the anterior, medial, ventral lateral, and posterior medial thalamus showed stronger evoked responses in response to higher working memory load. Activity in most thalamic nuclei were only modulated by working memory load, but not by categorical membership of the memorized stimuli, suggesting that thalamic function supports domain-general processing for working memory. To determine whether thalamocortical interactions contribute to cortical activity for working memory, we employed an activity flow mapping analysis to test whether thalamocortical interactions can predict cortical task activity patterns. In support, this data-driven thalamocortical interaction model explained a significant amount of variance in the observed cortical activity patterns modulated by working memory load. Our results suggest that the anterior, medial, and posterior medial thalamus, and their associated thalamocortical interactions, contribute to the modulations of distributed cortical activity during working memory.

Zoom-In Neural Network Deep-Learning Model for Alzheimer's Disease Assessments.

Deep neural networks have been successfully applied to generate predictive patterns from medical and diagnostic data. This paper presents an approach for assessing persons with Alzheimer's disease (AD) mild cognitive impairment (MCI), compared with normal control (NC) persons, using the zoom-in neural network (ZNN) deep-learning algorithm. ZNN stacks a set of zoom-in learning units (ZLUs) in a feedforward hierarchy without backpropagation. The resting-state fMRI (rs-fMRI) dataset for AD assessments was obtained from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The Automated Anatomical Labeling (AAL-90) atlas, which provides 90 neuroanatomical functional regions, was used to assess and detect the implicated regions in the course of AD. The features of the ZNN are extracted from the 140-time series rs-fMRI voxel values in a region of the brain. ZNN yields the three classification accuracies of AD versus MCI and NC, NC versus AD and MCI, and MCI versus AD and NC of 97.7%, 84.8%, and 72.7%, respectively, with the seven discriminative regions of interest (ROIs) in the AAL-90.