Two neuroanatomical subtypes of males with autism spectrum disorder revealed using semi-supervised machine learning

Three autism subtypes based on single-subject gray matter network revealed by semi-supervised machine learning

Autism spectrum disorder (ASD) is a heterogeneous, early-onset neurodevelopmental condition characterized by persistent impairments in social interaction and communication. This study aims to delineate ASD subtypes based on individual gray matter brain networks and provide new insights from a graph theory perspective. In this study, we extracted and normalized single-subject gray matter networks and calculated each network's topological properties. The heterogeneity through discriminative analysis (HYDRA) method was utilized to subtype all patients based on network properties. Next, we explored the differences among ASD subtypes in terms of network properties and clinical measures. Our investigation identified three distinct ASD subtypes. In the case-control study, these subtypes exhibited significant differences, particularly in the precentral gyrus, lingual gyrus, and middle frontal gyrus. In the case analysis, significant differences in global and nodal properties were observed between any two subtypes. Clinically, subtype 1 showed lower VIQ and PIQ compared to subtype 3, but exhibited higher scores in ADOS-Communication and ADOS-Total compared to subtype 2. The results highlight the distinct brain network properties and behaviors among different subtypes of male patients with ASD, providing valuable insights into the neural mechanisms underlying ASD heterogeneity.

Two multimodal neuroimaging subtypes of obsessive-compulsive disorder disclosed by semi-supervised machine learning

BACKGROUND

Obsessive-compulsive disorder (OCD) is a highly heterogeneous mental condition with a diverse symptom. Existing studies classified OCD on the basis of conventional phenomenology-based taxonomy ignoring the fact that the same subtype identified in accordance with clinical symptom may have different mechanisms and treatment responses.

METHODS

This research involved 50 medicine-free patients with OCD and 50 matched healthy controls (HCs). All the participants were subjected to structural and functional magnetic resonance imaging (MRI). Voxel-based morphometry (VBM) and amplitude of low frequency fluctuation (ALFF) were used to evaluate gray matter volume (GMV) and spontaneous neuronal activities at rest respectively. Similarity network fusion (SNF) was utilized to integrate GMVs and spontaneous neuronal activities, and heterogeneity by discriminant analysis was applied to characterise OCD subtypes.

RESULTS

Two OCD subtypes were identified: Subtype 1 exhibited decreased GMVs (i.e., left inferior temporal gyrus, right supplementary motor area and right lingual gyrus) and increased ALFF value (i.e., right orbitofrontal cortex), whereas subtype 2 exhibited increased GMVs (i.e., left cuneus, right precentral gyrus, left postcentral gyrus and left hippocampus) and decreased ALFF value (i.e., right caudate nucleus). Furthermore, the altered GMVs was negatively correlated with abnormal ALFF values in both subtype 1 and 2.

LIMITATIONS

This study requires further validation via a larger, independent dataset and should consider the potential influences of psychotropic medication on OCD patients' brain activities.

CONCLUSIONS

Results revealed two reproducible subtypes of OCD based on underlying multimodal neuroimaging and provided new perspectives on the classification of OCD.

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes with different brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal magnetic resonance imaging to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, and multiple sclerosis, as well as their potential in a transdiagnostic framework, where neuroanatomical and neurobiological commonalities were assessed across diagnostic boundaries. Subsequently, we summarize relevant machine learning methodologies and their clinical interpretability. We discuss the potential clinical implications of the current findings and envision future research avenues. Finally, we discuss an emerging paradigm called dimensional neuroimaging endophenotypes. Dimensional neuroimaging endophenotypes dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into low-dimensional yet informative, quantitative brain phenotypic representations, serving as robust intermediate phenotypes (i.e., endophenotypes), presumably reflecting the interplay of underlying genetic, lifestyle, and environmental processes associated with disease etiology.

Elucidating Multimodal Imaging Patterns in Accelerated Brain Aging: Heterogeneity through a Discriminant Analysis Approach Using the UK Biobank Dataset

Accelerated brain aging (ABA) intricately links with age-associated neurodegenerative and neuropsychiatric diseases, emphasizing the critical need for a nuanced exploration of heterogeneous ABA patterns. This investigation leveraged data from the UK Biobank (UKB) for a comprehensive analysis, utilizing structural magnetic resonance imaging (sMRI), diffusion magnetic resonance imaging (dMRI), and resting-state functional magnetic resonance imaging (rsfMRI) from 31,621 participants. Pre-processing employed tools from the FMRIB Software Library (FSL, version 5.0.10), FreeSurfer, DTIFIT, and MELODIC, seamlessly integrated into the UKB imaging processing pipeline. The Lasso algorithm was employed for brain-age prediction, utilizing derived phenotypes obtained from brain imaging data. Subpopulations of accelerated brain aging (ABA) and resilient brain aging (RBA) were delineated based on the error between actual age and predicted brain age. The ABA subgroup comprised 1949 subjects (experimental group), while the RBA subgroup comprised 3203 subjects (control group). Semi-supervised heterogeneity through discriminant analysis (HYDRA) refined and characterized the ABA subgroups based on distinctive neuroimaging features. HYDRA systematically stratified ABA subjects into three subtypes: SubGroup 2 exhibited extensive gray-matter atrophy, distinctive white-matter patterns, and unique connectivity features, displaying lower cognitive performance; SubGroup 3 demonstrated minimal atrophy, superior cognitive performance, and higher physical activity; and SubGroup 1 occupied an intermediate position. This investigation underscores pronounced structural and functional heterogeneity in ABA, revealing three subtypes and paving the way for personalized neuroprotective treatments for age-related neurological, neuropsychiatric, and neurodegenerative diseases.

Dimensional Neuroimaging Endophenotypes: Neurobiological Representations of Disease Heterogeneity Through Machine Learning

Machine learning has been increasingly used to obtain individualized neuroimaging signatures for disease diagnosis, prognosis, and response to treatment in neuropsychiatric and neurodegenerative disorders. Therefore, it has contributed to a better understanding of disease heterogeneity by identifying disease subtypes that present significant differences in various brain phenotypic measures. In this review, we first present a systematic literature overview of studies using machine learning and multimodal MRI to unravel disease heterogeneity in various neuropsychiatric and neurodegenerative disorders, including Alzheimer's disease, schizophrenia, major depressive disorder, autism spectrum disorder, multiple sclerosis, as well as their potential in transdiagnostic settings. Subsequently, we summarize relevant machine learning methodologies and discuss an emerging paradigm which we call dimensional neuroimaging endophenotype (DNE). DNE dissects the neurobiological heterogeneity of neuropsychiatric and neurodegenerative disorders into a low-dimensional yet informative, quantitative brain phenotypic representation, serving as a robust intermediate phenotype (i.e., endophenotype) largely reflecting underlying genetics and etiology. Finally, we discuss the potential clinical implications of the current findings and envision future research avenues.

Deep learning based joint fusion approach to exploit anatomical and functional brain information in autism spectrum disorders

BACKGROUND

The integration of the information encoded in multiparametric MRI images can enhance the performance of machine-learning classifiers. In this study, we investigate whether the combination of structural and functional MRI might improve the performances of a deep learning (DL) model trained to discriminate subjects with Autism Spectrum Disorders (ASD) with respect to typically developing controls (TD).

MATERIAL AND METHODS

We analyzed both structural and functional MRI brain scans publicly available within the ABIDE I and II data collections. We considered 1383 male subjects with age between 5 and 40 years, including 680 subjects with ASD and 703 TD from 35 different acquisition sites. We extracted morphometric and functional brain features from MRI scans with the Freesurfer and the CPAC analysis packages, respectively. Then, due to the multisite nature of the dataset, we implemented a data harmonization protocol. The ASD vs. TD classification was carried out with a multiple-input DL model, consisting in a neural network which generates a fixed-length feature representation of the data of each modality (FR-NN), and a Dense Neural Network for classification (C-NN). Specifically, we implemented a joint fusion approach to multiple source data integration. The main advantage of the latter is that the loss is propagated back to the FR-NN during the training, thus creating informative feature representations for each data modality. Then, a C-NN, with a number of layers and neurons per layer to be optimized during the model training, performs the ASD-TD discrimination. The performance was evaluated by computing the Area under the Receiver Operating Characteristic curve within a nested 10-fold cross-validation. The brain features that drive the DL classification were identified by the SHAP explainability framework.

RESULTS

The AUC values of 0.66±0.05 and of 0.76±0.04 were obtained in the ASD vs. TD discrimination when only structural or functional features are considered, respectively. The joint fusion approach led to an AUC of 0.78±0.04. The set of structural and functional connectivity features identified as the most important for the two-class discrimination supports the idea that brain changes tend to occur in individuals with ASD in regions belonging to the Default Mode Network and to the Social Brain.

CONCLUSIONS

Our results demonstrate that the multimodal joint fusion approach outperforms the classification results obtained with data acquired by a single MRI modality as it efficiently exploits the complementarity of structural and functional brain information.

Two distinct neuroanatomical subtypes of migraine without aura revealed by heterogeneity through discriminative analysis

The neurobiological heterogeneity in migraine is poorly studied, resulting in conflicting neuroimaging findings. This study used a newly proposed method based on gray matter volumes (GMVs) to investigate objective neuroanatomical subtypes of migraine. Structural MRI and clinical measures of 31 migraine patients without aura and 33 matched healthy controls (HCs) were explored. Firstly, we investigated whether migraine patients exhibited higher interindividual variability than HCs in terms of GMVs. Then, heterogeneity through discriminative analysis (HYDRA) was applied to categorize migraine patients into distinct subtypes by regional volumetric measures of GMVs. Voxel-wise volume and clinical characteristics among different subtypes were also explored. Migraine patients without aura exhibited higher interindividual GMVs variability. Two distinct and reproducible neuroanatomical subtypes of migraine were revealed. These two subtypes exhibited opposite neuroanatomical aberrances compared to HCs. Subtype 1 showed widespread decreased GMVs, while Subtype 2 showed increased GMVs in limited regions. The total intracranial volume was significantly positively correlated with cognitive function in Subtype 2. Subtype 1 showed significantly longer illness duration and less cognitive scores compared to Subtype 2. The present study shows that migraine patients without aura have high structural heterogeneity and uncovers two distinct and robust neuroanatomical subtypes, which provide a possible explanation for conflicting neuroimaging findings.

Two neurosubtypes of ADHD different from the clinical phenotypes

Clinical and etiological variability of attention deficit hyperactivity disorder (ADHD) presents an obstacle to understand the disorder. The aim of this study was to disentangle the heterogeneity of ADHD using neuroimaging and a semi-supervised machine learning algorithm. We collected brain structural and functional magnetic resonance imaging (MRI) data and clinical profiles of 183 children with ADHD and 396 neurotypical controls from 7 independent sites. We also used an external validation set with 750 subjects. We adopted a semi-supervised clustering method to subtype ADHD by regional volumetric measures of gray matter, white matter, and fractional amplitude of low frequency fluctuation (fALFF). In addition, split sample test, leave-one-site-out test and external validation were applied to evaluate the reproducibility and stability of ADHD subtypes. Two stable and reproducible neurosubtypes of ADHD were disclosed, which were proved by the split-sample test and leave-one-site-out validation. The structural and functional patterns of ADHD subtypes were also stable in the external validation set. The current two neurosubtypes differed in clinical manifestations and volumetric gray matter, white matter volume and fALFF patterns. The current neurosubtypes of ADHD which were different from clinical phenotypes could facilitate understanding the underlying neuropathological and neurobiological mechanism of the disorder.

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.

Parsing the heterogeneity of brain-symptom associations in autism spectrum disorder via random forest with homogeneous canonical correlation

BACKGROUND

Autism spectrum disorder (ASD) is a highly heterogeneous developmental disorder, but the neuroimaging substrates of its heterogeneity remain unknown. The difficulty lies mainly on the significant individual variability in the brain-symptom association.

METHODS

T1-weighted magnetic resonance imaging data from the Autism Brain Imaging Database Exchange (ABIDE) (N_TDC = 1146) were used to generate a normative model to map brain structure deviations of cases (N_ASD = 571). Voxel-based morphometry (VBM) was used to compute gray matter volume (GMV). Singular Value Decomposition (SVD) was employed to perform dimensionality reduction. A tree-based algorithm was proposed to identify the ASD subtypes according to the pattern of brain-symptom association as assessed by a homogeneous canonical correlation.

RESULTS

We identified 4 ASD subtypes with distinct association patterns between residual volumes and a social symptom score. More severe the social symptom was associated with greater GMVs in both the frontoparietal regions for the subtype1 (r = 0.29-0.44) and the ventral visual pathway for the subtype3 (r = 0.19-0.23), but lower GMVs in both the right anterior cingulate cortex for the subtype4 (r = -0.25) and a few subcortical regions for the subtype2 (r = -0.31-0.20). The subtyping significantly improved the classification accuracy between cases and controls from 0.64 to 0.75 (p < 0.05, permutation test), which was also better than the accuracy of 0.68 achieved by the k-means-based subtyping (p < 0.01).

LIMITATIONS

Sample size limited the study due to the missing data.

CONCLUSIONS

These findings suggest that the heterogeneity of ASD might reflect changes in different subsystems of the social brain, especially including social attention, motivation, perceiving and evaluation.

Current Approaches in Computational Psychiatry for the Data-Driven Identification of Brain-Based Subtypes

The ability of our current psychiatric nosology to accurately delineate clinical populations and inform effective treatment plans has reached a critical point with only moderately successful interventions and high relapse rates. These challenges continue to motivate the search for approaches to better stratify clinical populations into more homogeneous delineations, to better inform diagnosis and disease evaluation, and prescribe and develop more precise treatment plans. The promise of brain-based subtyping based on neuroimaging data is that finding subgroups of individuals with a common biological signature will facilitate the development of biologically grounded, targeted treatments. This review provides a snapshot of the current state of the field in empirical brain-based subtyping studies in child, adolescent, and adult psychiatric populations published between 2019 and March 2022. We found that there is vast methodological exploration and a surprising number of new methods being created for the specific purpose of brain-based subtyping. However, this methodological exploration and advancement is not being met with rigorous validation approaches that assess both reproducibility and clinical utility of the discovered brain-based subtypes. We also found evidence for a collaboration crisis, in which methodological exploration and advancements are not clearly grounded in clinical goals. We propose several steps that we believe are crucial to address these shortcomings in the field. We conclude, and agree with the authors of the reviewed studies, that the discovery of biologically grounded subtypes would be a significant advancement for treatment development in psychiatry.