Association of polygenic scores for autism with volumetric MRI phenotypes in cerebellum and brainstem in adults.

Previous research on autism spectrum disorders (ASD) have showed important volumetric alterations in the cerebellum and brainstem. Most of these studies are however limited to case-control studies with small clinical samples and including mainly children or adolescents. Herein, we aimed to explore the association between the cumulative genetic load (polygenic risk score, PRS) for ASD and volumetric alterations in the cerebellum and brainstem, as well as global brain tissue volumes of the brain among adults at the population level. We utilized the latest genome-wide association study of ASD by the Psychiatric Genetics Consortium (18,381 cases, 27,969 controls) and constructed the ASD PRS in an independent cohort, the UK Biobank. Regression analyses controlled for multiple comparisons with the false-discovery rate (FDR) at 5% were performed to investigate the association between ASD PRS and forty-four brain magnetic resonance imaging (MRI) phenotypes among ~ 31,000 participants. Primary analyses included sixteen MRI phenotypes: total volumes of the brain, cerebrospinal fluid (CSF), grey matter (GM), white matter (WM), GM of whole cerebellum, brainstem, and ten regions of the cerebellum (I_IV, V, VI, VIIb, VIIIa, VIIIb, IX, X, CrusI and CrusII). Secondary analyses included twenty-eight MRI phenotypes: the sub-regional volumes of cerebellum including the GM of the vermis and both left and right lobules of each cerebellar region. ASD PRS were significantly associated with the volumes of seven brain areas, whereby higher PRS were associated to reduced volumes of the whole brain, WM, brainstem, and cerebellar regions I-IV, IX, and X, and an increased volume of the CSF. Three sub-regional volumes including the left cerebellar lobule I-IV, cerebellar vermes VIIIb, and X were significantly and negatively associated with ASD PRS. The study highlights a substantial connection between susceptibility to ASD, its underlying genetic etiology, and neuroanatomical alterations of the adult brain.

The diagnosis of ASD with MRI: a systematic review and meta-analysis.

While diagnosing autism spectrum disorder (ASD) based on an objective test is desired, the current diagnostic practice involves observation-based criteria. This study is a systematic review and meta-analysis of studies that aim to diagnose ASD using magnetic resonance imaging (MRI). The main objective is to describe the state of the art of diagnosing ASD using MRI in terms of performance metrics and interpretation. Furthermore, subgroups, including different MRI modalities and statistical heterogeneity, are analyzed. Studies that dichotomously diagnose individuals with ASD and healthy controls by analyses progressing from magnetic resonance imaging obtained in a resting state were systematically selected by two independent reviewers. Studies were sought on Web of Science and PubMed, which were last accessed on February 24, 2023. The included studies were assessed on quality and risk of bias using the revised Quality Assessment of Diagnostic Accuracy Studies tool. A bivariate random-effects model was used for syntheses. One hundred and thirty-four studies were included comprising 159 eligible experiments. Despite the overlap in the studied samples, an estimated 4982 unique participants consisting of 2439 individuals with ASD and 2543 healthy controls were included. The pooled summary estimates of diagnostic performance are 76.0% sensitivity (95% CI 74.1-77.8), 75.7% specificity (95% CI 74.0-77.4), and an area under curve of 0.823, but uncertainty in the study assessments limits confidence. The main limitations are heterogeneity and uncertainty about the generalization of diagnostic performance. Therefore, comparisons between subgroups were considered inappropriate. Despite the current limitations, methods progressing from MRI approach the diagnostic performance needed for clinical practice. The state of the art has obstacles but shows potential for future clinical application.

Associations of alcohol and tobacco use with psychotic, depressive and developmental disorders revealed via multimodal neuroimaging.

People affected by psychotic, depressive and developmental disorders are at a higher risk for alcohol and tobacco use. However, the further associations between alcohol/tobacco use and symptoms/cognition in these disorders remain unexplored. We identified multimodal brain networks involving alcohol use (n = 707) and tobacco use (n = 281) via supervised multimodal fusion and evaluated if these networks affected symptoms and cognition in people with psychotic (schizophrenia/schizoaffective disorder/bipolar, n = 178/134/143), depressive (major depressive disorder, n = 260) and developmental (autism spectrum disorder/attention deficit hyperactivity disorder, n = 421/346) disorders. Alcohol and tobacco use scores were used as references to guide functional and structural imaging fusion to identify alcohol/tobacco use associated multimodal patterns. Correlation analyses between the extracted brain features and symptoms or cognition were performed to evaluate the relationships between alcohol/tobacco use with symptoms/cognition in 6 psychiatric disorders. Results showed that (1) the default mode network (DMN) and salience network (SN) were associated with alcohol use, whereas the DMN and fronto-limbic network (FLN) were associated with tobacco use; (2) the DMN and fronto-basal ganglia (FBG) related to alcohol/tobacco use were correlated with symptom and cognition in psychosis; (3) the middle temporal cortex related to alcohol/tobacco use was associated with cognition in depression; (4) the DMN related to alcohol/tobacco use was related to symptom, whereas the SN and limbic system (LB) were related to cognition in developmental disorders. In summary, alcohol and tobacco use were associated with structural and functional abnormalities in DMN, SN and FLN and had significant associations with cognition and symptoms in psychotic, depressive and developmental disorders likely via different brain networks. Further understanding of these relationships may assist clinicians in the development of future approaches to improve symptoms and cognition among psychotic, depressive and developmental disorders.

Association between Autism Spectrum Disorder, Trace Elements, and Intracranial Fluid Spaces.

(1) Autism spectrum disorder (ASD) belongs to the group of complex developmental disorders. Novel studies have suggested that genetic and environmental factors equally affect the risk of ASD. Identification of environmental factors involved in the development of ASD is therefore crucial for a better understanding of its etiology. Whether there is a causal link between trace elements, brain magnetic resonance imaging (MRI), and ASD remains a matter of debate and requires further studies. (2) In the prospective part of the study, we included 194 children, including an age-matched control group; in the retrospective study, 28 children with available MRI imaging were included. All children had urine analysis of trace elements performed. In those with available brain MRI, linear indexes for the ventricular volumes were measured and calculated. (3) We found the highest vanadium, rubidium, thallium, and silver levels in children with ASD. These elements also correlated with the estimated ventricular volume based on MRI indexes in children with ASD in the subanalysis. However, the severity of the deficits did not correlate with brain MRI indexes of our elements, except negatively with magnesium. (4) Trace elements have an impact on children with ASD, but further multi-centric studies are needed to explain the pathophysiological mechanisms.

Resolving autism spectrum disorder (ASD) through brain topologies using fMRI dataset with multi-layer perceptron (MLP).

Autism is a neurodevelopmental disorder that manifests in individuals during childhood and has enduring consequences for their social interactions and communication. The prediction of Autism Spectrum Disorder (ASD) in individuals based on the differences in brain networks and activities have been studied extensively in the recent past, however, with lower accuracies. Therefore in this research, identification at the early stage through computer-aided algorithms to differentiate between ASD and TD patients is proposed. In order to identify features, a Multi-Layer Perceptron (MLP) model is developed which utilizes logistic regression on characteristics extracted from connectivity matrices of subjects derived from fMRI images. The features that significantly contribute to the classification of individuals as having Autism Spectrum Disorder (ASD) or typically developing (TD) are identified by the logistic regression model. To enhance emphasis on essential attributes, an AND operation is integrated. This involves selecting features demonstrating statistical significance across diverse logistic regression analyses conducted on various random distributions. The iterative approach contributes to a comprehensive understanding of relevant features for accurate classification. By implementing this methodology, the estimation of feature importance became more dependable, and the potential for overfitting is moderated through the evaluation of model performance on various subsets of data. It is observed from the experimentation that the highly correlated Left Lateral Occipital Cortex and Right Lateral Occipital Cortex ROIs are only found in ASD. Also, it is noticed that the highly correlated Left Cerebellum Tonsil and Right Cerebellum Tonsil are only found in TD participants. Among the MLP classifier, a recall of 82.61 % is achieved followed by Logistic Regression with an accuracy of 72.46 %. MLP also stands out with a commendable accuracy of 83.57 % and AUC of 0.978.

Mapping neural correlates of biological motion perception in autistic children using high-density diffuse optical tomography.

BACKGROUND: Autism spectrum disorder (ASD), a neurodevelopmental disorder defined by social communication deficits plus repetitive behaviors and restricted interests, currently affects 1/36 children in the general population. Recent advances in functional brain imaging show promise to provide useful biomarkers of ASD diagnostic likelihood, behavioral trait severity, and even response to therapeutic intervention. However, current gold-standard neuroimaging methods (e.g., functional magnetic resonance imaging, fMRI) are limited in naturalistic studies of brain function underlying ASD-associated behaviors due to the constrained imaging environment. Compared to fMRI, high-density diffuse optical tomography (HD-DOT), a non-invasive and minimally constraining optical neuroimaging modality, can overcome these limitations. Herein, we aimed to establish HD-DOT to evaluate brain function in autistic and non-autistic school-age children as they performed a biological motion perception task previously shown to yield results related to both ASD diagnosis and behavioral traits. METHODS: We used HD-DOT to image brain function in 46 ASD school-age participants and 49 non-autistic individuals (NAI) as they viewed dynamic point-light displays of coherent biological and scrambled motion. We assessed group-level cortical brain function with statistical parametric mapping. Additionally, we tested for brain-behavior associations with dimensional metrics of autism traits, as measured with the Social Responsiveness Scale-2, with hierarchical regression models. RESULTS: We found that NAI participants presented stronger brain activity contrast (coherent > scrambled) than ASD children in cortical regions related to visual, motor, and social processing. Additionally, regression models revealed multiple cortical regions in autistic participants where brain function is significantly associated with dimensional measures of ASD traits. LIMITATIONS: Optical imaging methods are limited in depth sensitivity and so cannot measure brain activity within deep subcortical regions. However, the field of view of this HD-DOT system includes multiple brain regions previously implicated in both task-based and task-free studies on autism. CONCLUSIONS: This study demonstrates that HD-DOT is sensitive to brain function that both differentiates between NAI and ASD groups and correlates with dimensional measures of ASD traits. These findings establish HD-DOT as an effective tool for investigating brain function in autistic and non-autistic children. Moreover, this study established neural correlates related to biological motion perception and its association with dimensional measures of ASD traits.

A multi-frequency approach of the altered functional connectome for autism spectrum disorder identification.

Autism spectrum disorder stands as a multifaceted and heterogeneous neurodevelopmental condition. The utilization of functional magnetic resonance imaging to construct functional brain networks proves instrumental in comprehending the intricate interplay between brain activity and autism spectrum disorder, thereby elucidating the underlying pathogenesis at the cerebral level. Traditional functional brain networks, however, typically confine their examination to connectivity effects within a specific frequency band, disregarding potential connections among brain areas that span different frequency bands. To harness the full potential of interregional connections across diverse frequency bands within the brain, our study endeavors to develop a novel multi-frequency analysis method for constructing a comprehensive functional brain networks that incorporates multiple frequencies. Specifically, our approach involves the initial decomposition of functional magnetic resonance imaging into distinct frequency bands through wavelet transform. Subsequently, Pearson correlation is employed to generate corresponding functional brain networks and kernel for each frequency band. Finally, the classification was performed by a multi-kernel support vector machine, to preserve the connectivity effects within each band and the connectivity patterns shared among the different bands. Our proposed multi-frequency functional brain networks method yielded notable results, achieving an accuracy of 89.1%, a sensitivity of 86.67%, and an area under the curve of 0.942 in a publicly available autism spectrum disorder dataset.

Environmental enrichment reduces restricted repetitive behavior by altering gray matter microstructure.

Restricted, repetitive behaviors are common symptoms in neurodevelopmental disorders including autism spectrum disorder. Despite being associated with poor developmental outcomes, repetitive behaviors remain poorly understood and have limited treatment options. Environmental enrichment attenuates the development of repetitive behaviors, but the exact mechanisms remain obscure. Using the C58 mouse model of repetitive behavior, we performed diffusion tensor imaging to examine microstructural alterations associated with the development of repetitive behavior and its attenuation by environmental enrichment. The C57BL/6 mouse strain, which displays little or no repetitive behavior, was used as a control group. We observed widespread differences in diffusion metrics between C58 mice and C57BL/6 mice. In juvenile C58 mice, repetitive motor behavior displayed strong negative correlations with fractional anisotropy in multiple gray matter regions, whereas in young adult C58 mice, high repetitive motor behavior was most strongly associated with lower fractional anisotropy and higher radial diffusivity in the striatum. Environmental enrichment increased fractional anisotropy and axial diffusivity throughout gray matter regions in the brains of juvenile C58 mice and overlapped predominantly with cerebellar and sensory regions associated with repetitive behavior. Our results suggest environmental enrichment reduces repetitive behavior development by altering gray matter microstructure in the cerebellum, medial entorhinal cortex, and sensory processing regions in juvenile C58 mice. Under standard laboratory conditions, early pathology in these regions appears to contribute to later striatal and white matter dysfunction in adult C58 mice. Future studies should examine the role these regions play in the development of repetitive behavior and the relationship between sensory processing and cerebellar deficits and repetitive behavior.

Disentangling the role of gray matter volume and concentration in autism spectrum disorder: A meta-analytic investigation of 25 years of voxel-based morphometry research.

Despite over two decades of neuroimaging research, a unanimous definition of the pattern of structural variation associated with autism spectrum disorder (ASD) has yet to be found. One potential impeding issue could be the sometimes ambiguous use of measurements of variations in gray matter volume (GMV) or gray matter concentration (GMC). In fact, while both can be calculated using voxel-based morphometry analysis, these may reflect different underlying pathological mechanisms. We conducted a coordinate-based meta-analysis, keeping apart GMV and GMC studies of subjects with ASD. Results showed distinct and non-overlapping patterns for the two measures. GMV decreases were evident in the cerebellum, while GMC decreases were mainly found in the temporal and frontal regions. GMV increases were found in the parietal, temporal, and frontal brain regions, while GMC increases were observed in the anterior cingulate cortex and middle frontal gyrus. Age-stratified analyses suggested that such variations are dynamic across the ASD lifespan. The present findings emphasize the importance of considering GMV and GMC as distinct yet synergistic indices in autism research.

Exploring autism via the retina: Comparative insights in children with autism spectrum disorder and typical development.

Autism spectrum disorder (ASD) is a widely recognized neurodevelopmental disorder, yet the identification of reliable imaging biomarkers for its early diagnosis remains a challenge. Considering the specific manifestations of ASD in the eyes and the interconnectivity between the brain and the eyes, this study investigates ASD through the lens of retinal analysis. We specifically examined differences in the macular region of the retina using optical coherence tomography (OCT)/optical coherence tomography angiography (OCTA) images between children diagnosed with ASD and those with typical development (TD). Our findings present potential novel characteristics of ASD: the thickness of the ellipsoid zone (EZ) with cone photoreceptors was significantly increased in ASD; the large-caliber arteriovenous of the inner retina was significantly reduced in ASD; these changes in the EZ and arteriovenous were more significant in the left eye than in the right eye. These observations of photoreceptor alterations, vascular function changes, and lateralization phenomena in ASD warrant further investigation, and we hope that this work can advance interdisciplinary understanding of ASD.

Altered white matter connectivity of ventral language networks in autism spectrum disorder: An automated fiber quantification analysis with multi-site datasets.

Comprehension and pragmatic deficits are prevalent in autism spectrum disorder (ASD) and are potentially linked to altered connectivity in the ventral language networks. However, previous magnetic resonance imaging studies have not sufficiently explored the microstructural abnormalities in the ventral fiber tracts underlying comprehension dysfunction in ASD. Additionally, the precise locations of white matter (WM) changes in the long tracts of patients with ASD remain poorly understood. In the current study, we applied the automated fiber-tract quantification (AFQ) method to investigate the fine-grained WM properties of the ventral language pathway and their relationships with comprehension and symptom manifestation in ASD. The analysis included diffusion/T1 weighted imaging data of 83 individuals with ASD and 83 age-matched typically developing (TD) controls. Case-control comparisons were performed on the diffusion metrics of the ventral tracts at both the global and point-wise levels. We also explored correlations between diffusion metrics, comprehension performance, and ASD traits, and conducted subgroup analyses based on age range to examine developmental moderating effects. Individuals with ASD exhibited remarkable hypoconnectivity in the ventral tracts, particularly in the temporal portions of the left inferior longitudinal fasciculus (ILF) and the inferior fronto-occipital fasciculus (IFOF). These WM abnormalities were associated with poor comprehension and more severe ASD symptoms. Furthermore, WM alterations in the ventral tract and their correlation with comprehension dysfunction were more prominent in younger children with ASD than in adolescents. These findings indicate that WM disruptions in the temporal portions of the left ILF/IFOF are most notable in ASD, potentially constituting the core neurological underpinnings of comprehension and communication deficits in autism. Moreover, impaired WM connectivity and comprehension ability in patients with ASD appear to improve with age.

Brain functional gradient and structure features in adolescent and adult autism spectrum disorders.

Understanding how function and structure are organized and their coupling with clinical traits in individuals with autism spectrum disorder (ASD) is a primary goal in network neuroscience research for ASD. Atypical brain functional networks and structures in individuals with ASD have been reported, but whether these associations show heterogeneous hierarchy modeling in adolescents and adults with ASD remains to be clarified. In this study, 176 adolescent and 74 adult participants with ASD without medication or comorbidities and sex, age matched healthy controls (HCs) from 19 research groups from the openly shared Autism Brain Imaging Data Exchange II database were included. To investigate the relationship between the functional gradient, structural changes, and clinical symptoms of brain networks in adolescents and adults with ASD, functional gradient and voxel-based morphometry (VBM) analyses based on 1000 parcels defined by Schaefer mapped to Yeo's seven-network atlas were performed. Pearson's correlation was calculated between the gradient scores, gray volume and density, and clinical traits. The subsystem-level analysis showed that the second gradient scores of the default mode networks and frontoparietal network in patients with ASD were relatively compressed compared to adolescent HCs. Adult patients with ASD showed an overall compression gradient of 1 in the ventral attention networks. In addition, the gray density and volumes of the subnetworks showed no significant differences between the ASD and HC groups at the adolescent stage. However, adults with ASD showed decreased gray density in the limbic network. Moreover, numerous functional gradient parameters, but not VBM parameters, in adolescents with ASD were considerably correlated with clinical traits in contrast to those in adults with ASD. Our findings proved that the atypical changes in adolescent ASD mainly involve the brain functional network, while in adult ASD, the changes are more related to brain structure, including gray density and volume. These changes in functional gradients or structures are markedly correlated with clinical traits in patients with ASD. Our study provides a novel understanding of the pathophysiology of the structure-function hierarchy in ASD.

Sex-related patterns of functional brain networks in children and adolescents with autism spectrum disorder.

Although numerous studies have emphasized the male predominance in autism spectrum disorder (ASD), how sex differences are related to the topological organization of functional networks remains unclear. This study utilized imaging data from 86 ASD (43 females, aged 7-18 years) and 86 typically developing controls (TCs) (43 females, aged 7-18 years) obtained from Autism Brain Imaging Data Exchange databases, constructed individual whole-brain functional networks, used a graph theory analysis to compute topological metrics, and assessed sex-related differences in topological metrics using a 2 × 2 factorial design. At the global level, females with ASD exhibited significantly higher cluster coefficient and local efficiency than female TCs, while no significant difference was observed between males with ASD and male TCs. Meanwhile, the neurotypical sex differences in cluster coefficient and local efficiency observed in TCs were not present in ASD. At the nodal level, ASD exhibited abnormal nodal centrality in the left middle temporal gyrus.

Exploring Implicit Biological Heterogeneity in ASD Diagnosis Using a Multi-Head Attention Graph Neural Network.

BACKGROUND: Autism spectrum disorder (ASD) is a neurodevelopmental disorder exhibiting heterogeneous characteristics in patients, including variability in developmental progression and distinct neuroanatomical features influenced by sex and age. Recent advances in deep learning models based on functional connectivity (FC) graphs have produced promising results, but they have focused on generalized global activation patterns and failed to capture specialized regional characteristics and accurately assess disease indications. METHODS: To overcome these limitations, we propose a novel deep learning method that models FC with multi-head attention, which enables simultaneous modeling of the intricate and variable patterns of brain connectivity associated with ASD, effectively extracting abnormal patterns of brain connectivity. The proposed method not only identifies region-specific correlations but also emphasizes connections at specific, transient time points from diverse perspectives. The extracted FC is transformed into a graph, assigning weighted labels to the edges to reflect the degree of correlation, which is then processed using a graph neural network capable of handling edge labels. RESULTS: Experiments on the autism brain imaging data exchange (ABIDE) I and II datasets, which include a heterogeneous cohort, showed superior performance over the state-of-the-art methods, improving accuracy by up to 3.7%p. The incorporation of multi-head attention in FC analysis markedly improved the distinction between typical brains and those affected by ASD. Additionally, the ablation study validated diverse brain characteristics in ASD patients across different ages and sexes, offering insightful interpretations. CONCLUSION: These results emphasize the effectiveness of the method in enhancing diagnostic accuracy and its potential in advancing neurological research for ASD diagnosis.

Advancing ASD identification with neuroimaging: a novel GARL methodology integrating Deep Q-Learning and generative adversarial networks.

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition that affects an individual's behavior, speech, and social interaction. Early and accurate diagnosis of ASD is pivotal for successful intervention. The limited availability of large datasets for neuroimaging investigations, however, poses a significant challenge to the timely and precise identification of ASD. To address this problem, we propose a breakthrough approach, GARL, for ASD diagnosis using neuroimaging data. GARL innovatively integrates the power of GANs and Deep Q-Learning to augment limited datasets and enhance diagnostic precision. We utilized the Autistic Brain Imaging Data Exchange (ABIDE) I and II datasets and employed a GAN to expand these datasets, creating a more robust and diversified dataset for analysis. This approach not only captures the underlying sample distribution within ABIDE I and II but also employs deep reinforcement learning for continuous self-improvement, significantly enhancing the capability of the model to generalize and adapt. Our experimental results confirmed that GAN-based data augmentation effectively improved the performance of all prediction models on both datasets, with the combination of InfoGAN and DQN's GARL yielding the most notable improvement.

Gender and age related brain structural and functional alterations in children with autism spectrum disorder.

To explore the effects of age and gender on the brain in children with autism spectrum disorder using magnetic resonance imaging. 185 patients with autism spectrum disorder and 110 typically developing children were enrolled. In terms of gender, boys with autism spectrum disorder had increased gray matter volumes in the insula and superior frontal gyrus and decreased gray matter volumes in the inferior frontal gyrus and thalamus. The brain regions with functional alterations are mainly distributed in the cerebellum, anterior cingulate gyrus, postcentral gyrus, and putamen. Girls with autism spectrum disorder only had increased gray matter volumes in the right cuneus and showed higher amplitude of low-frequency fluctuation in the paracentral lobule, higher regional homogeneity and degree centrality in the calcarine fissure, and greater right frontoparietal network-default mode network connectivity. In terms of age, preschool-aged children with autism spectrum disorder exhibited hypo-connectivity between and within auditory network, somatomotor network, and visual network. School-aged children with autism spectrum disorder showed increased gray matter volumes in the rectus gyrus, superior temporal gyrus, insula, and suboccipital gyrus, as well as increased amplitude of low-frequency fluctuation and regional homogeneity in the calcarine fissure and precentral gyrus and decreased in the cerebellum and anterior cingulate gyrus. The hyper-connectivity between somatomotor network and left frontoparietal network and within visual network was found. It is essential to consider the impact of age and gender on the neurophysiological alterations in autism spectrum disorder children when analyzing changes in brain structure and function.

Altered intra- and inter-network connectivity in autism spectrum disorder.

OBJECTIVE: A neurodevelopmental illness termed as the autism spectrum disorder (ASD) is described by social interaction impairments. Previous studies employing resting-state functional imaging (rs-fMRI) identified both hyperconnectivity and hypoconnectivity patterns in ASD people. However, specific patterns of connectivity within and between networks linked to ASD remain largely unexplored. METHODS: We utilized a meticulously selected subset of high-quality data, comprising 45 individuals diagnosed with ASD and 47 HCs, obtained from the ABIDE dataset. The pre-processed rs-fMRI time series signals were partitioned into ninety regions of interest. We focused on eight intrinsic connectivity networks and further performed intra- and inter-network analysis. Finally, support vector machine was used to discriminate ASD from HC. RESULTS: Through different sparsities, ASD exhibited significantly decreased intra-network connectivity within default mode network and dorsal attention network, increased connectivity between limbic network and subcortical network, and decreased connectivity between default mode network and limbic network. Using the classifier trained on altered intra- and inter-network connectivity, multivariate pattern analyses classified the ASD from HC with 71.74% accuracy, 70.21% specificity and 75.56% sensitivity in 10% sparsity of functional connectivity. CONCLUSIONS: ASD showed characteristic reorganization of the brain networks and this provided new insight into the underlying process of the functional connectome dysfunction in ASD.

Exploring the potential of representation and transfer learning for anatomical neuroimaging: Application to psychiatry.

The perspective of personalized medicine for brain disorders requires efficient learning models for anatomical neuroimaging-based prediction of clinical conditions. There is now a consensus on the benefit of deep learning (DL) in addressing many medical imaging tasks, such as image segmentation. However, for single-subject prediction problems, recent studies yielded contradictory results when comparing DL with Standard Machine Learning (SML) on top of classical feature extraction. Most existing comparative studies were limited in predicting phenotypes of little clinical interest, such as sex and age, and using a single dataset. Moreover, they conducted a limited analysis of the employed image pre-processing and feature selection strategies. This paper extensively compares DL and SML prediction capacity on five multi-site problems, including three increasingly complex clinical applications in psychiatry namely schizophrenia, bipolar disorder, and Autism Spectrum Disorder (ASD) diagnosis. To compensate for the relative scarcity of neuroimaging data on these clinical datasets, we also evaluate three pre-training strategies for transfer learning from brain imaging of the general healthy population: self-supervised learning, generative modeling and supervised learning with age. Overall, we find similar performance between randomly initialized DL and SML for the three clinical tasks and a similar scaling trend for sex prediction. This was replicated on an external dataset. We also show highly correlated discriminative brain regions between DL and linear ML models in all problems. Nonetheless, we demonstrate that self-supervised pre-training on large-scale healthy population imaging datasets (N≈10k), along with Deep Ensemble, allows DL to learn robust and transferable representations to smaller-scale clinical datasets (N≤1k). It largely outperforms SML on 2 out of 3 clinical tasks both in internal and external test sets. These findings suggest that the improvement of DL over SML in anatomical neuroimaging mainly comes from its capacity to learn meaningful and useful abstract representations of the brain anatomy, and it sheds light on the potential of transfer learning for personalized medicine in psychiatry.

ASD-SWNet: a novel shared-weight feature extraction and classification network for autism spectrum disorder diagnosis.

The traditional diagnostic process for autism spectrum disorder (ASD) is subjective, where early and accurate diagnosis significantly affects treatment outcomes and life quality. Thus, improving ASD diagnostic methods is critical. This paper proposes ASD-SWNet, a new shared-weight feature extraction and classification network. It resolves the issue found in previous studies of inefficiently integrating unsupervised and supervised learning, thereby enhancing diagnostic precision. The approach utilizes functional magnetic resonance imaging to improve diagnostic accuracy, featuring an autoencoder (AE) with Gaussian noise for robust feature extraction and a tailored convolutional neural network (CNN) for classification. The shared-weight mechanism utilizes features learned by the AE to initialize the convolutional layer weights of the CNN, thereby integrating AE and CNN for joint training. A novel data augmentation strategy for time-series medical data is also introduced, tackling the problem of small sample sizes. Tested on the ABIDE-I dataset through nested ten-fold cross-validation, the method achieved an accuracy of 76.52% and an AUC of 0.81. This approach surpasses existing methods, showing significant enhancements in diagnostic accuracy and robustness. The contribution of this paper lies not only in proposing new methods for ASD diagnosis but also in offering new approaches for other neurological brain diseases.