Machine learning for autism spectrum disorder diagnosis using structural magnetic resonance imaging: Promising but challenging

Enhanced detection of autism spectrum disorder through neuroimaging data using stack classifier ensembled with modified VGG-19

BackgroundAutism spectrum disorder (ASD) is a neurodevelopmental disease marked by a variety of repetitive behaviors and social communication difficulties.PurposeTo develop a generalizable machine learning (ML) classifier that can accurately and effectively predict ASD in children.Material and MethodsThis paper makes use of neuroimaging data from the Autism Brain Imaging Data Exchange (ABIDE I and II) datasets through a combination of structural and functional magnetic resonance imaging data. Several ML models, such as Support Vector Machines (SVM), CatBoost, random forest (RF), and stack classifiers, were tested to demonstrate which model performs the best in ASD classification when used alongside a deep convolutional neural network.ResultsResults showed that stack classifier performed the best among the models, with the highest accuracy of 81.68%, sensitivity of 85.08%, and specificity of 79.13% for ABIDE I, and 81.34%, 83.61%, and 82.21% for ABIDE II, showing its superior ability to identify complex patterns in neuroimaging data. SVM performed poorly across all metrics, showing its limitations in dealing with high-dimensional neuroimaging data.ConclusionThe results show that the application of ML models, especially ensemble approaches like stack classifier, holds significant promise in improving the accuracy with which ASD is detected using neuroimaging and thus shows their potential for use in clinical applications and early intervention strategies.

AI-based non-invasive imaging technologies for early autism spectrum disorder diagnosis: A short review and future directions

Autism Spectrum Disorder (ASD) is a neurological condition, with recent statistics from the CDC indicating a rising prevalence of ASD diagnoses among infants and children. This trend emphasizes the critical importance of early detection, as timely diagnosis facilitates early intervention and enhances treatment outcomes. Consequently, there is an increasing urgency for research to develop innovative tools capable of accurately and objectively identifying ASD in its earliest stages. This paper offers a short overview of recent advancements in non-invasive technology for early ASD diagnosis, focusing on an imaging modality, structural MRI technique, which has shown promising results in early ASD diagnosis. This brief review aims to address several key questions: (i) Which imaging radiomics are associated with ASD? (ii) Is the parcellation step of the brain cortex necessary to improve the diagnostic accuracy of ASD? (iii) What databases are available to researchers interested in developing non-invasive technology for ASD? (iv) How can artificial intelligence tools contribute to improving the diagnostic accuracy of ASD? Finally, our review will highlight future trends in ASD diagnostic efforts.

Latest clinical frontiers related to autism diagnostic strategies

The diagnosis of autism is currently based on the developmental history, direct observation of behavior, and reported symptoms, supplemented by rating scales/interviews/structured observational evaluations-which is influenced by the clinician's knowledge and experience-with no established diagnostic biomarkers. A growing body of research has been conducted over the past decades to improve diagnostic accuracy. Here, we provide an overview of the current diagnostic assessment process as well as of recent and ongoing developments to support diagnosis in terms of genetic evaluation, telemedicine, digital technologies, use of machine learning/artificial intelligence, and research on candidate diagnostic biomarkers. Genetic testing can meaningfully contribute to the assessment process, but caution is required when interpreting negative results, and more work is needed to strengthen the transferability of genetic information into clinical practice. Digital diagnostic and machine-learning-based analyses are emerging as promising approaches, but larger and more robust studies are needed. To date, there are no available diagnostic biomarkers. Moving forward, international collaborations may help develop multimodal datasets to identify biomarkers, ensure reproducibility, and support clinical translation.

Mapping the structure of biomarkers in autism spectrum disorder: a review of the most influential studies

BACKGROUND

Autism spectrum disorder is a distinctive developmental condition which is caused by an interaction between genetic vulnerability and environmental factors. Biomarkers play a crucial role in understanding disease characteristics for diagnosis, prognosis, and treatment. This study employs bibliometric analysis to identify and review the 100 top-cited articles' characteristics, current research hotspots and future directions of autism biomarkers.

METHODS

A comprehensive search of autism biomarkers studies was retrieved from the Web of Science Core Collection database with a combined keyword search strategy. A comprehensive analysis of the top 100 articles was conducted with CiteSpace, VOSviewer, and Excel, including citations, countries, authors, and keywords.

RESULTS

The top 100 cited studies were published between 1988 and 2021, with the United States led in productivity. Core biomarkers such as genetics, children, oxidative stress, and mitochondrial dysfunction are well-established. Potential trends for future research may include brain studies, metabolomics, and associations with other psychiatric disorders.

CONCLUSION

This pioneering bibliometric analysis provides a comprehensive compilation of the 100 most-cited studies on autism, which not only offers a valuable resource for doctors, and researchers but shedding insights into current shortcomings and future endeavors. Future research should prioritize the application of emerging technologies for biomarkers, longitudinal study of biomarkers, and specificity of autism biomarkers to advance the precision of ASD diagnosis and treatment.

Transparent deep learning to identify autism spectrum disorders (ASD) in EHR using clinical notes

OBJECTIVE

Machine learning (ML) is increasingly employed to diagnose medical conditions, with algorithms trained to assign a single label using a black-box approach. We created an ML approach using deep learning that generates outcomes that are transparent and in line with clinical, diagnostic rules. We demonstrate our approach for autism spectrum disorders (ASD), a neurodevelopmental condition with increasing prevalence.

METHODS

We use unstructured data from the Centers for Disease Control and Prevention (CDC) surveillance records labeled by a CDC-trained clinician with ASD A1-3 and B1-4 criterion labels per sentence and with ASD cases labels per record using Diagnostic and Statistical Manual of Mental Disorders (DSM5) rules. One rule-based and three deep ML algorithms and six ensembles were compared and evaluated using a test set with 6773 sentences (N = 35 cases) set aside in advance. Criterion and case labeling were evaluated for each ML algorithm and ensemble. Case labeling outcomes were compared also with seven traditional tests.

RESULTS

Performance for criterion labeling was highest for the hybrid BiLSTM ML model. The best case labeling was achieved by an ensemble of two BiLSTM ML models using a majority vote. It achieved 100% precision (or PPV), 83% recall (or sensitivity), 100% specificity, 91% accuracy, and 0.91 F-measure. A comparison with existing diagnostic tests shows that our best ensemble was more accurate overall.

CONCLUSIONS

Transparent ML is achievable even with small datasets. By focusing on intermediate steps, deep ML can provide transparent decisions. By leveraging data redundancies, ML errors at the intermediate level have a low impact on final outcomes.

A hybrid CNN-SVM model for enhanced autism diagnosis

Autism is a representative disorder of pervasive developmental disorder. It exerts influence upon an individual's behavior and performance, potentially co-occurring with other mental illnesses. Consequently, an effective diagnostic approach proves to be invaluable in both therapeutic interventions and the timely provision of medical support. Currently, most scholars' research primarily relies on neuroimaging techniques for auxiliary diagnosis and does not take into account the distinctive features of autism's social impediments. In order to address this deficiency, this paper introduces a novel convolutional neural network-support vector machine model that integrates resting state functional magnetic resonance imaging data with the social responsiveness scale metrics for the diagnostic assessment of autism. We selected 821 subjects containing the social responsiveness scale measure from the publicly available Autism Brain Imaging Data Exchange dataset, including 379 subjects with autism spectrum disorder and 442 typical controls. After preprocessing of fMRI data, we compute the static and dynamic functional connectivity for each subject. Subsequently, convolutional neural networks and attention mechanisms are utilized to extracts their respective features. The extracted features, combined with the social responsiveness scale features, are then employed as novel inputs for the support vector machine to categorize autistic patients and typical controls. The proposed model identifies salient features within the static and dynamic functional connectivity, offering a possible biological foundation for clinical diagnosis. By incorporating the behavioral assessments, the model achieves a remarkable classification accuracy of 94.30%, providing a more reliable support for auxiliary diagnosis.

Machine learning algorithms to the early diagnosis of fetal alcohol spectrum disorders

Introduction

Fetal alcohol spectrum disorders include a variety of physical and neurocognitive disorders caused by prenatal alcohol exposure. Although their overall prevalence is around 0.77%, FASD remains underdiagnosed and little known, partly due to the complexity of their diagnosis, which shares some symptoms with other pathologies such as autism spectrum, depression or hyperactivity disorders.

Methods

This study included 73 control and 158 patients diagnosed with FASD. Variables selected were based on IOM classification from 2016, including sociodemographic, clinical, and psychological characteristics. Statistical analysis included Kruskal-Wallis test for quantitative factors, Chi-square test for qualitative variables, and Machine Learning (ML) algorithms for predictions.

Results

This study explores the application ML in diagnosing FASD and its subtypes: Fetal Alcohol Syndrome (FAS), partial FAS (pFAS), and Alcohol-Related Neurodevelopmental Disorder (ARND). ML constructed a profile for FASD based on socio-demographic, clinical, and psychological data from children with FASD compared to a control group. Random Forest (RF) model was the most efficient for predicting FASD, achieving the highest metrics in accuracy (0.92), precision (0.96), sensitivity (0.92), F1 Score (0.94), specificity (0.92), and AUC (0.92). For FAS, XGBoost model obtained the highest accuracy (0.94), precision (0.91), sensitivity (0.91), F1 Score (0.91), specificity (0.96), and AUC (0.93). In the case of pFAS, RF model showed its effectiveness, with high levels of accuracy (0.90), precision (0.86), sensitivity (0.96), F1 Score (0.91), specificity (0.83), and AUC (0.90). For ARND, RF model obtained the best levels of accuracy (0.87), precision (0.76), sensitivity (0.93), F1 Score (0.84), specificity (0.83), and AUC (0.88). Our study identified key variables for efficient FASD screening, including traditional clinical characteristics like maternal alcohol consumption, lip-philtrum, microcephaly, height and weight impairment, as well as neuropsychological variables such as the Working Memory Index (WMI), aggressive behavior, IQ, somatic complaints, and depressive problems.

Discussion

Our findings emphasize the importance of ML analyses for early diagnoses of FASD, allowing a better understanding of FASD subtypes to potentially improve clinical practice and avoid misdiagnosis.

Etiologic Evaluation of Children with Autism Spectrum Disorder

Autism spectrum disorder (ASD) is clinically and etiologically heterogeneous. A causal genetic variant can be identified in approximately 20% to 25% of affected individuals with current clinical genetic testing, and all patients with an ASD diagnosis should be offered genetic etiologic evaluation. We suggest that exome sequencing with copy number variant coverage should be the first-line etiologic evaluation for ASD. Neuroimaging, neurophysiologic, metabolic, and other biochemical evaluations can provide insight into the pathophysiology of ASD but should be recommended in the appropriate clinical circumstances.

The Dynamical Biomarkers in Functional Connectivity of Autism Spectrum Disorder Based on Dynamic Graph Embedding

Autism spectrum disorder (ASD) is a neurological and developmental disorder and its early diagnosis is a challenging task. The dynamic brain network (DBN) offers a wealth of information for the diagnosis and treatment of ASD. Mining the spatio-temporal characteristics of DBN is critical for finding dynamic communication across brain regions and, ultimately, identifying the ASD diagnostic biomarker. We proposed the dgEmbed-KNN and the Aggregation-SVM diagnostic models, which use the spatio-temporal information from DBN and interactive information among brain regions represented by dynamic graph embedding. The classification accuracies show that dgEmbed-KNN model performs slightly better than traditional machine learning and deep learning methods, while the Aggregation-SVM model has a very good capacity to diagnose ASD using aggregation brain network connections as features. We discovered over- and under-connections in ASD at the level of dynamic connections, involving brain regions of the postcentral gyrus, the insula, the cerebellum, the caudate nucleus, and the temporal pole. We also found abnormal dynamic interactions associated with ASD within/between the functional subnetworks, including default mode network, visual network, auditory network and saliency network. These can provide potential DBN biomarkers for ASD identification.

Validation of plasma protein glycation and oxidation biomarkers for the diagnosis of autism

Autism Spectrum Disorder (ASD) is a common neurodevelopmental disorder in children. It is currently diagnosed by behaviour-based assessments made by observation and interview. In 2018 we reported a discovery study of a blood biomarker diagnostic test for ASD based on a combination of four plasma protein glycation and oxidation adducts. The test had 88% accuracy in children 5-12 years old. Herein, we present an international multicenter clinical validation study (N = 478) with application of similar biomarkers to a wider age range of 1.5-12 years old children. Three hundred and eleven children with ASD (247 male, 64 female; age 5.2 ± 3.0 years) and 167 children with typical development (94 male, 73 female; 4.9 ± 2.4 years) were recruited for this study at Sidra Medicine and Hamad Medical Corporation hospitals, Qatar, and Hospital Regional Universitario de Málaga, Spain. For subjects 5-12 years old, the diagnostic algorithm with features, advanced glycation endproducts (AGEs)-Nε-carboxymethyl-lysine (CML), Nω-carboxymethylarginine (CMA) and 3-deoxyglucosone-derived hydroimidazolone (3DG-H), and oxidative damage marker, o,o'-dityrosine (DT), age and gender had accuracy 83% (CI 79 - 89%), sensitivity 94% (CI 90-98%), specificity 67% (CI 57-76%) and area-under-the-curve of receiver operating characteristic plot (AUROC) 0.87 (CI 0.84-0.90). Inclusion of additional plasma protein glycation and oxidation adducts increased the specificity to 74%. An algorithm with 12 plasma protein glycation and oxidation adduct features was optimum for children of 1.5-12 years old: accuracy 74% (CI 70-79%), sensitivity 75% (CI 63-87%), specificity 74% (CI 58-90%) and AUROC 0.79 (CI 0.74-0.84). We conclude that ASD diagnosis may be supported using an algorithm with features of plasma protein CML, CMA, 3DG-H and DT in 5-12 years-old children, and an algorithm with additional features applicable for ASD screening in younger children. ASD severity, as assessed by ADOS-2 score, correlated positively with plasma protein glycation adducts derived from methylglyoxal, hydroimidazolone MG-H1 and Nε(1-carboxyethyl)lysine (CEL). The successful validation herein may indicate that the algorithm modifiable features are mechanistic risk markers linking ASD to increased lipid peroxidation, neuronal plasticity and proteotoxic stress.

Diagnosing autism severity associated with physical fitness and gray matter volume in children with autism spectrum disorder: Explainable machine learning method

PURPOSE

This study aimed to investigate the relationship between physical fitness, gray matter volume (GMV), and autism severity in children with autism spectrum disorder (ASD). Besides, we sought to diagnose autism severity associated with physical fitness and GMV using machine learning methods.

METHODS

Ninety children diagnosed with ASD underwent physical fitness tests, magnetic resonance imaging scans, and autism severity assessments. Diagnosis models were established using extreme gradient boosting (XGB), random forest (RF), support vector machine (SVM), and decision tree (DT) algorithms. Hyperparameters were optimized through the grid search cross-validation method. The shapley additive explanation (SHAP) method was employed to explain the diagnosis results.

RESULTS

Our study revealed associations between muscular strength in physical fitness and GMV in specific brain regions (left paracentral lobule, bilateral thalamus, left inferior temporal gyrus, and cerebellar vermis I-II) with autism severity in children with ASD. The accuracy (95 % confidence interval) of the XGB, RF, SVM, and DT models were 77.9 % (77.3, 78.6 %), 72.4 % (71.7, 73.2 %), 71.9 % (71.1, 72.6 %), and 66.9 % (66.2, 67.7 %), respectively. SHAP analysis revealed that muscular strength and thalamic GMV significantly influenced the decision-making process of the XGB model.

CONCLUSION

Machine learning methods can effectively diagnose autism severity associated with physical fitness and GMV in children with ASD. In this respect, the XGB model demonstrated excellent performance across various indicators, suggesting its potential for diagnosing autism severity.

Comprehensive exploration of multi-modal and multi-branch imaging markers for autism diagnosis and interpretation: insights from an advanced deep learning model

Autism spectrum disorder is a complex neurodevelopmental condition with diverse genetic and brain involvement. Despite magnetic resonance imaging advances, autism spectrum disorder diagnosis and understanding its neurogenetic factors remain challenging. We propose a dual-branch graph neural network that effectively extracts and fuses features from bimodalities, achieving 73.9% diagnostic accuracy. To explain the mechanism distinguishing autism spectrum disorder from healthy controls, we establish a perturbation model for brain imaging markers and perform a neuro-transcriptomic joint analysis using partial least squares regression and enrichment to identify potential genetic biomarkers. The perturbation model identifies brain imaging markers related to structural magnetic resonance imaging in the frontal, temporal, parietal, and occipital lobes, while functional magnetic resonance imaging markers primarily reside in the frontal, temporal, occipital lobes, and cerebellum. The neuro-transcriptomic joint analysis highlights genes associated with biological processes, such as "presynapse," "behavior," and "modulation of chemical synaptic transmission" in autism spectrum disorder's brain development. Different magnetic resonance imaging modalities offer complementary information for autism spectrum disorder diagnosis. Our dual-branch graph neural network achieves high accuracy and identifies abnormal brain regions and the neuro-transcriptomic analysis uncovers important genetic biomarkers. Overall, our study presents an effective approach for assisting in autism spectrum disorder diagnosis and identifying genetic biomarkers, showing potential for enhancing the diagnosis and treatment of this condition.

Morphometric network-based abnormalities correlate with psychiatric comorbidities and gene expression in PCDH19-related developmental and epileptic encephalopathy

Protocadherin-19 (PCDH19) developmental and epileptic encephalopathy causes an early-onset epilepsy syndrome with limbic seizures, typically occurring in clusters and variably associated with intellectual disability and a range of psychiatric disorders including hyperactive, obsessive-compulsive and autistic features. Previous quantitative neuroimaging studies revealed abnormal cortical areas in the limbic formation (parahippocampal and fusiform gyri) and underlying white-matter fibers. In this study, we adopted morphometric, network-based and multivariate statistical methods to examine the cortex and substructure of the hippocampus and amygdala in a cohort of 20 PCDH19-mutated patients and evaluated the relation between structural patterns and clinical variables at individual level. We also correlated morphometric alterations with known patterns of PCDH19 expression levels. We found patients to exhibit high-significant reductions of cortical surface area at a whole-brain level (left/right pvalue = 0.045/0.084), and particularly in the regions of the limbic network (left/right parahippocampal gyri pvalue = 0.230/0.016; left/right entorhinal gyri pvalue = 0.002/0.327), and bilateral atrophy of several subunits of the amygdala and hippocampus, particularly in the CA regions (head of the left CA3 pvalue = 0.002; body of the right CA3 pvalue = 0.004), and differences in the shape of hippocampal structures. More severe psychiatric comorbidities correlated with more significant altered patterns, with the entorhinal gyrus (pvalue = 0.013) and body of hippocampus (pvalue = 0.048) being more severely affected. Morphometric alterations correlated significantly with the known expression patterns of PCDH19 (rvalue = -0.26, pspin = 0.092). PCDH19 encephalopathy represents a model of genetically determined neural network based neuropsychiatric disease in which quantitative MRI-based findings correlate with the severity of clinical manifestations and had have a potential predictive value if analyzed early.

Do it the transformer way: A comprehensive review of brain and vision transformers for autism spectrum disorder diagnosis and classification

Autism spectrum disorder (ASD) is a condition observed in children who display abnormal patterns of interaction, behavior, and communication with others. Despite extensive research efforts, the underlying causes of this neurodevelopmental disorder and its biomarkers remain unknown. However, advancements in artificial intelligence and machine learning have improved clinicians' ability to diagnose ASD. This review paper investigates various MRI modalities to identify distinct features that characterize individuals with ASD compared to typical control subjects. The review then moves on to explore deep learning models for ASD diagnosis, including convolutional neural networks (CNNs), autoencoders, graph convolutions, attention networks, and other models. CNNs and their variations are particularly effective due to their capacity to learn structured image representations and identify reliable biomarkers for brain disorders. Computer vision transformers often employ CNN architectures with transfer learning techniques like fine-tuning and layer freezing to enhance image classification performance, surpassing traditional machine learning models. This review paper contributes in three main ways. Firstly, it provides a comprehensive overview of a recommended architecture for using vision transformers in the systematic ASD diagnostic process. To this end, the paper investigates various pre-trained vision architectures such as VGG, ResNet, Inception, InceptionResNet, DenseNet, and Swin models that were fine-tuned for ASD diagnosis and classification. Secondly, it discusses the vision transformers of 2020th like BiT, ViT, MobileViT, and ConvNeXt, and applying transfer learning methods in relation to their prospective practicality in ASD classification. Thirdly, it explores brain transformers that are pre-trained on medically rich data and MRI neuroimaging datasets. The paper recommends a systematic architecture for ASD diagnosis using brain transformers. It also reviews recently developed brain transformer-based models, such as METAFormer, Com-BrainTF, Brain Network, ST-Transformer, STCAL, BolT, and BrainFormer, discussing their deep transfer learning architectures and results in ASD detection. Additionally, the paper summarizes and discusses brain-related transformers for various brain disorders, such as MSGTN, STAGIN, and MedTransformer, in relation to their potential usefulness in ASD. The study suggests that developing specialized transformer-based models, following the success of natural language processing (NLP), can offer new directions for image classification problems in ASD brain biomarkers learning and classification. By incorporating the attention mechanism, treating MRI modalities as sequence prediction tasks trained on brain disorder classification problems, and fine-tuned on ASD datasets, brain transformers can show a great promise in ASD diagnosis.

Few-Shot Classification of Autism Spectrum Disorder using Site-Agnostic Meta-Learning and Brain MRI

For machine learning applications in medical imaging, the availability of training data is often limited, which hampers the design of radiological classifiers for subtle conditions such as autism spectrum disorder (ASD). Transfer learning is one method to counter this problem of low training data regimes. Here we explore the use of meta-learning for very low data regimes in the context of having prior data from multiple sites - an approach we term site-agnostic meta-learning. Inspired by the effectiveness of meta-learning for optimizing a model across multiple tasks, here we propose a framework to adapt it to learn across multiple sites. We tested our meta-learning model for classifying ASD versus typically developing controls in 2,201 T1-weighted (T1-w) MRI scans collected from 38 imaging sites as part of Autism Brain Imaging Data Exchange (ABIDE) [age: 5.2 -64.0 years]. The method was trained to find a good initialization state for our model that can quickly adapt to data from new unseen sites by fine-tuning on the limited data that is available. The proposed method achieved an area under the receiver operating characteristic curve (ROC-AUC)=0.857 on 370 scans from 7 unseen sites in ABIDE using a few-shot setting of 2-way 20-shot i.e., 20 training samples per site. Our results outperformed a transfer learning baseline by generalizing across a wider range of sites as well as other related prior work. We also tested our model in a zero-shot setting on an independent test site without any additional fine-tuning. Our experiments show the promise of the proposed site-agnostic meta-learning framework for challenging neuroimaging tasks involving multi-site heterogeneity with limited availability of training data.Clinical Relevance- We propose a learning framework that accommodates multi-site heterogeneity and limited data to assist in challenging neuroimaging tasks.

Applications of Deep Learning to Neurodevelopment in Pediatric Imaging: Achievements and Challenges

Deep learning has achieved remarkable progress, particularly in neuroimaging analysis. Deep learning applications have also been extended from adult to pediatric medical images, and thus, this paper aims to present a systematic review of this recent research. We first introduce the commonly used deep learning methods and architectures in neuroimaging, such as convolutional neural networks, auto-encoders, and generative adversarial networks. A non-exhaustive list of commonly used publicly available pediatric neuroimaging datasets and repositories are included, followed by a categorical review of recent works in pediatric MRI-based deep learning studies in the past five years. These works are categorized into recognizing neurodevelopmental disorders, identifying brain and tissue structures, estimating brain age/maturity, predicting neurodevelopment outcomes, and optimizing MRI brain imaging and analysis. Finally, we also discuss the recent achievements and challenges on these applications of deep learning to pediatric neuroimaging.