EXPERIMENTAL ANALYSIS OF ASD DETECTION - COMPARING CNN VARIANTS AND HYBRID CNN-RNN WITH AAL MAP

Within the light of the DSM-5, the current study examined (1) how many and which children with a DSM-IV classification of autism spectrum disorder (ASD) fulfill the DSM-5 symptom-criteria, and (2) whether children who did and did not meet DSM-5 symptom-criteria and children with social anxiety disorder (SAD) can be differentiated from each other based on ASD symptomatology. In total, 90 referred children with a DSM-IV classification of high-functioning ASD, and 21 referred children with SAD participated (age range 7–17 years). ASD-symptoms were examined with the Autism Diagnostic Interview-Revised and the Children’s Social Behavior Questionnaire. It was found that 30 % of the ASD sample did not meet DSM-5 symptom-criteria for ASD, mainly because they failed to meet the DSM-5 criteria of the repetitive domain. Children with ASD who did and did not meet DSM-5 symptom-criteria differed on the repetitive domain, while children with ASD (according to DSM-IV and DSM-5 symptom criteria) had higher scores on the social-communication domain than children with SAD. Findings suggest a continuum of ASD-symptoms in the DSM-5 for children with SAD, social communication disorder and ASD. More research is needed to examine how these three disorders differ with respect to their etiology, neuropsychological profiles and clinical characteristics.

BackgroundGenetic studies have implicated rare and common variations in liability for autism spectrum disorder (ASD). Of the discovered risk variants, those rare in the population invariably have large impact on liability, while common variants have small effects. Yet, collectively, common risk variants account for the majority of population-level variability. How these rare and common risk variants jointly affect liability for individuals requires further study.MethodsTo explore how common and rare variants jointly affect liability, we assessed two cohorts of ASD families characterized for rare and common genetic variations (Simons Simplex Collection and Population-Based Autism Genetics and Environment Study). We analyzed data from 3011 affected subjects, as well as two cohorts of unaffected individuals characterized for common genetic variation: 3011 subjects matched for ancestry to ASD subjects and 11,950 subjects for estimating allele frequencies. We used genetic scores, which assessed the relative burden of common genetic variation affecting risk of ASD (henceforth “burden”), and determined how this burden was distributed among three subpopulations: ASD subjects who carry a potentially damaging variant implicated in risk of ASD (“PDV carriers”); ASD subjects who do not (“non-carriers”); and unaffected subjects who are assumed to be non-carriers.ResultsBurden harbored by ASD subjects is stochastically greater than that harbored by control subjects. For PDV carriers, their average burden is intermediate between non-carrier ASD and control subjects. Both carrier and non-carrier ASD subjects have greater burden, on average, than control subjects. The effects of common and rare variants likely combine additively to determine individual-level liability.LimitationsOnly 305 ASD subjects were known PDV carriers. This relatively small subpopulation limits this study to characterizing general patterns of burden, as opposed to effects of specific PDVs or genes. Also, a small fraction of subjects that are categorized as non-carriers could be PDV carriers.ConclusionsLiability arising from common and rare risk variations likely combines additively to determine risk of any individual diagnosed with ASD. On average, ASD subjects carry a substantial burden of common risk variation, even if they also carry a rare PDV affecting risk.

Autism Spectrum Disorder (ASD) is a multifactorial disorder involving genetic and environmental factors leading to pathophysiologic symptoms and comorbidities including neurodevelopmental disorders, anxiety, immune dysregulation, and gastrointestinal (GI) abnormalities. Abnormal intestinal permeability has been reported among ASD patients and it is well established that disturbances in eating patterns may cause gut microbiome imbalance (i.e., dysbiosis). Therefore, studies focusing on the potential relationship between gut microbiota and ASD are emerging. We compared the intestinal bacteriome and mycobiome of a cohort of ASD subjects with their non-ASD siblings. Differences between ASD and non-ASD subjects include a significant decrease at the phylum level in Cyanobacteria (0.015% vs. 0.074%, p < 0.0003), and a significant decrease at the genus level in Bacteroides (28.3% vs. 36.8%, p < 0.03). Species-level analysis showed a significant decrease in Faecalibacterium prausnitzii, Prevotella copri, Bacteroides fragilis, and Akkermansia municiphila. Mycobiome analysis showed an increase in the fungal Ascomycota phylum (98.3% vs. 94%, p < 0.047) and an increase in Candida albicans (27.1% vs. 13.2%, p < 0.055). Multivariate analysis showed that organisms from the genus Delftia were predictive of an increased odds ratio of ASD, whereas decreases at the phylum level in Cyanobacteria and at the genus level in Azospirillum were associated with an increased odds ratio of ASD. We screened 24 probiotic organisms to identify strains that could alter the growth patterns of organisms identified as elevated within ASD subject samples. In a preliminary in vivo preclinical test, we challenged wild-type Balb/c mice with Delftia acidovorans (increased in ASD subjects) by oral gavage and compared changes in behavioral patterns to sham-treated controls. An in vitro biofilm assay was used to determine the ability of potentially beneficial microorganisms to alter the biofilm-forming patterns of Delftia acidovorans, as well as their ability to break down fiber. Downregulation of cyanobacteria (generally beneficial for inflammation and wound healing) combined with an increase in biofilm-forming species such as D. acidovorans suggests that ASD-related GI symptoms may result from decreases in beneficial organisms with a concomitant increase in potential pathogens, and that beneficial probiotics can be identified that counteract these changes.

Brain magnetic resonance imaging (MRI) studies suggest the prevalence of asymptomatic "incidental" findings (IF) in autism spectrum disorder (ASD) is similar to that of neurotypically developing (NT) controls. However, given the causes of IF may include both genetic and environmental factors, a twin study would facilitate comparing brain IF between ASD and NT subjects. MRI scans were examined to assess the prevalence of brain IF in twin "case pairs" (at least one twin with diagnosis of ASD) and twin "control pairs" (NT). Fifty case pairs and thirty-two control pairs were analyzed. IF were found in 68% of subjects with ASD, 71% of unaffected ASD siblings, and in 58% of control subjects (P = 0.4). IF requiring clinical follow-up occurred more frequently in subjects with ASD compared to NT controls (17% vs. 5%, respectively; P = 0.02). The concordance rate of IF in twins was 83%. A mixed effects model found younger age, male sex, and "family environment" to be significantly associated with IF. There was no difference in the prevalence rate of IF between ASD subjects and NT controls. More IF required clinical follow-up in ASD subjects compared to NT controls. The prevalence rate of IF observed in this twin study was higher than rates previously reported in singleton studies. Our results suggest the shared environment of twins - perhaps in utero - increases the risk of brain IF. Brain MRI in the initial work-up of ASD may be indicated in twins, especially in males. Autism Res 2017, 10: 113-120. © 2016 International Society for Autism Research, Wiley Periodicals, Inc.

A challenging goal in neuroscience is that of identifying specific brain patterns characterising autistic spectrum disorder (ASD). Genetic studies, together with investigations based on magnetic resonance imaging (MRI) and functional MRI, support the idea that distinctive structural features could exist in the ASD brain. In the developing brains of babies and small children, structural differences could provide the basis for different brain connectivity, giving rise to macroscopic effects detectable by e.g. electroencephalography (EEG). A significant body of research has already been conducted in this direction, mainly computing spectral power and coherence. Perhaps due to methodological limitations, together with high variability within and between the cohorts investigated, results have not been in complete agreement, and it is therefore still the case that the diagnosis of ASD is based on behavioural tests and interviews. This thesis describes a step-by-step characterisation and comparison of brain dynamics from ASD and neurotypical subjects, based on the analysis of multi-probe EEG time-series from male children aged 3-5 years. The methods applied are all ones that take explicit account of the intrinsically non-linear, open, and time-variable nature of the system. Time-frequency representations were first computed from the time-series to evaluate the spectral power and to categorise the ranges encompassing different activities as low-frequency (LF, 0.8-3.5 Hz), mid-range-frequency (MF, 3.5-12 Hz) or high-frequency (HF, 12-48 Hz). The spatial pathways for the propagation of neuronal activity were then investigated by calculation of wavelet phase coherence. Finally, deeper insight into brain connectivity was achieved by computation of the dynamical cross-frequency coupling between triplets of spatially distributed phases. In doing so, dynamical Bayesian inference was used to find the coupling parameters between the oscillators in the spatially-distributed network. The sets of parameters extracted by this means allowed evaluation of the strength of particular coupling components of the triplet LF, MF→HF, and enabled reconstruction of the coupling functions. By investigation of the form of the coupling functions, the thesis goes beyond conventional measures like the directionality and strength of an interaction, and reveals subtler features of the underlying mechanism. The measured power distributions highlight differences between ASD and typically developing children in the preferential frequency range for local synchronisation of neuronal activity: the relative power is generally higher at LF and HF, and lower at MF, in the ASD case. The phase coherence maps from ASD subjects also exhibited differences, with lower connectivity at LF and MF in the frontal and fronto-occipital pairs, and higher coherence at high frequencies for central links. There was higher inter-subject variability in a comparison of the forms of coupling functions in the ASD group; and a weaker coupling in their theta-gamma range, which can be linked with the cognitive features of the disorder. In conclusion, the approach developed in this thesis gave promising preliminary results, suggesting that a biomarker for ASD could be defined in terms of the described patterns of functional and effective connectivity computed from EEG measurements.

We recently described a unique plasma metabolite profile in subjects with pediatric acute-onset neuropsychiatric syndrome (PANS), suggesting pathogenic models involving specific patterns of neurotransmission, neuroinflammation, and oxidative stress. Here, we extend the analysis to a group of patients with autism spectrum disorder (ASD), as a consensus has recently emerged around its immune-mediated pathophysiology with a widespread involvement of brain networks. This observational case-control study enrolled patients referred for PANS and ASD from June 2019 to May 2020, as well as neurotypical age and gender-matched control subjects. Thirty-four PANS outpatients, fifteen ASD outpatients, and twenty-five neurotypical subjects underwent physical and neuropsychiatric evaluations, alongside serum metabolomic analysis with 1H-NMR. In supervised models, the metabolomic profile of ASD was significantly different from controls (p = 0.0001), with skewed concentrations of asparagine, aspartate, betaine, glycine, lactate, glucose, and pyruvate. Metabolomic separation was also observed between PANS and ASD subjects (p = 0.02), with differences in the concentrations of arginine, aspartate, betaine, choline, creatine phosphate, glycine, pyruvate, and tryptophan. We confirmed a unique serum metabolomic profile of PANS compared with both ASD and neurotypical subjects, distinguishing PANS as a pathophysiological entity per se. Tryptophan and glycine appear as neuroinflammatory fingerprints of PANS and ASD, respectively. In particular, a reduction in glycine would primarily affect NMDA-R excitatory tone, overall impairing downstream glutamatergic, dopaminergic, and GABAergic transmissions. Nonetheless, we found metabolomic similarities between PANS and ASD that suggest a putative role of N-methyl-D-aspartate receptor (NMDA-R) dysfunction in both disorders. Metabolomics-based approaches could contribute to the identification of novel ASD and PANS biomarkers.

Elevated expression of toll-like receptor 4 is associated with NADPH oxidase-induced oxidative stress in B cells of children with autism

Autism spectrum disorder (ASD) is a complex developmental disorder in children that results in abnormal communicative and verbal behaviors. Exposure to heavy metals plays a significant role in the pathogenesis or progression of ASD. Mercury compounds pose significant risk for the development of ASD as children are more exposed to environmental toxicants. Increased concentration of mercury compounds has been detected in different body fluids/tissues in ASD children, which suggests an association between mercury exposure and ASD. Thioredoxin1 (Trx1) and thioredoxin reductase1 (TrxR1) redox system plays a crucial role in detoxification of oxidants generated in different immune cells. However, the effect of methylmercury and the Nrf2 activator sulforaphane on the Trx1/TrxR1 antioxidant system in neutrophils of ASD subjects has not been studied previously. Therefore, this study examined the effect of methylmercury on Trx1/TrxR1 expression, TrxR activity, nitrotyrosine, and ROS in neutrophils of ASD and TDC subjects. Our study shows that Trx1/TrxR1 protein expression is dysregulated in ASD subjects as compared to the TDC group. Further, methylmercury treatment significantly inhibits the activity of TrxR in both ASD and TDC groups. Inhibition of TrxR by mercury is associated with upregulation of the Trx1 protein in TDC neutrophils but not in ASD neutrophils. Furthermore, ASD neutrophils have exaggerated ROS production after exposure to methylmercury, which is much greater in magnitude than TDC neutrophils. Sulforaphane reversed methylmercury-induced effects on neutrophils through Nrf2-mediated induction of the Trx1/TrxR1 system. These observations suggest that exposure to the environmental toxicant methylmercury may elevate systemic oxidative inflammation due to a dysregulated Trx1/TrxR1 redox system in the neutrophils of ASD subjects, which may play a role in the progression of ASD.

Autism spectrum disorder (ASD) encompasses a group of neurodevelopmental conditions diagnosed solely on the basis of behavioral assessments that reveal social deficits. Progress has been made in understanding its genetic underpinnings, but most ASD-associated genetic variants, which include copy number variants (CNVs) and mutations in ASD-risk genes, account for no more than 1 % of ASD cases. This high level of genetic heterogeneity leads to challenges obtaining and interpreting genetic testing in clinical settings. The traditional definition of syndromic ASD is a disorder with a clinically defined pattern of somatic abnormalities and a neurobehavioral phenotype that may include ASD. Most have a known genetic cause. Examples include fragile X syndrome and tuberous sclerosis complex. We propose dividing syndromic autism into the following two groups: (i) ASD that occurs in the context of a clinically defined syndrome-recognizing these disorders depends on the familiarity of the clinician with the features of the syndrome, and the diagnosis is typically confirmed by targeted genetic testing (eg, mutation screening of FMR1); (ii) ASD that occurs as a feature of a molecularly defined syndrome-for this group of patients, ASD-associated variants are identified by genome-wide testing that is not hypothesis driven (eg, microarray, whole exome sequencing). These ASD groups cannot be easily clinically defined because patients with a given variant have variable somatic abnormalities (dysmorphism and birth defects). In this article, we review common diagnoses from the above categories and suggest a testing strategy for patients, guided by determining whether the individual has essential or complex ASD; patients in the latter group have multiple morphologic anomalies on physical examination. Finally, we recommend that the syndromic versus nonsyndromic designation ultimately be replaced by classification of ASD according to its genetic etiology, which will inform about the associated spectrum and penetrance of neurobehavioral and somatic manifestations.

To examine the risk for premature mortality and intentional self-harm in autism spectrum disorders (ASD). Based on a national birth cohort. Children born in 1987–2005, diagnosed with ASD by 2007 (n = 4695) were matched with four non-ASD subjects (n = 18,450) and followed until 2015 for mortality and intentional self-harm. The risk among ASD subjects was elevated only for natural cause of death. The risk for intentional self-harm was increased in the unadjusted analyses, but decreased to non-significant after adjusting for comorbid psychiatric disorders. ASD subjects are at increased risk for premature mortality due to natural causes of death. While ASD subjects die of suicide and present with more self-harm, the association is explained by comorbid psychiatric disorders.

• Autism spectrum disorders (ASDs) comprise a group of neurodevelopmental disorders clinically characterized by social communication deficits and restricted and repetitive behaviors .• This condition is reported to affect 1 in 59 individuals according to the last Centers for diseases control and prevention (CDC) update of autism’s estimated prevalence.• Autism spectrum disorders (ASDs) is considered a formidable challenge for neuroscience because of their high prevalence, lifelong nature, complexity and substantial heterogeneity .• Although the etiology and pathogenesis of ASD are unclear ,the connectivity theory of ASD has been paid extensive attention in recent years. Functional magnetic resonance imaging (fMRI) has played an important role in studying the abnormal neurobiological underpinnings in ASD.• Resting-state fMRI has provided a convenient tool to examine the changes in the intrinsic connectivity of specific regions and networks in ASD and non-ASD subjects.• In this study, we further pursued to delineate patterns of intrinsic functional connectivity (iFC) in children with ASD and its relation to behavioral symptomatology, using rs-fMRI, hoping to shed another light of understanding on the complexity of ASD which may be a step forward for developing brain based biomarkers of the disorder.

Limited sample size hinders the application of deep learning in brain image analysis, and transfer learning is a possible solution. However, most pretrained models are 2D based and cannot be applied directly to 3D brain images. In this study, we propose a novel framework to apply 2D pretrained models to 3D brain images by projecting surface-based cortical morphometry into planar images using computational geometry mapping. Firstly, 3D cortical meshes are reconstructed from magnetic resonance imaging (MRI) using FreeSurfer and projected into 2D planar meshes with topological preservation based on area-preserving geometry mapping. Then, 2D deep models pretrained on ImageNet are adopted and fine-tuned for cortical image classification on morphometric shape metrics. We apply the framework to sex classification on the Human Connectome Project dataset and autism spectrum disorder (ASD) classification on the Autism Brain Imaging Data Exchange dataset. Moreover, a 2-stage transfer learning strategy is suggested to boost the ASD classification performance by using the sex classification as an intermediate task. Our framework brings significant improvement in sex classification and ASD classification with transfer learning. In summary, the proposed framework builds a bridge between 3D cortical data and 2D models, making 2D pretrained models available for brain image analysis in cognitive and psychiatric neuroscience.

Autism Spectrum Disorder (ASD) is a lifelong neurodevelopmental condition characterized by challenges in social communication and repetitive behaviors, making early diagnosis essential for effective intervention. This preliminary study examines the feasibility of classifying ASD using 2D slices extracted from 3D brain MRI volumes, focusing on the comparative diagnostic value of axial, coronal, and sagittal planes within a machine learning framework. The goal is to identify the most informative anatomical plane to guide future neuroimaging protocols and improve automated ASD detection. We analyzed 20,992 slices derived from 82 MRI scans (41 ASD and 41 non-ASD) from the Autism Brain Imaging Data Exchange (ABIDE), each scan averaging 256 × 256 pixels across 256 slices. Preprocessing included skull stripping and conversion to NIFTI-GZ format. Texture features were extracted using the Gray Level Co-occurrence Matrix (GLCM), and classification was performed using the K-Nearest Neighbors (KNN) algorithm with varying K values (3, 5, and 7). The best performance was achieved with K = 3 and GLCM parameters at distance d = 0 and angle 45°, yielding an accuracy of 83.90%, precision of 79%, recall of 78%, and F1-score of 78%. These findings provide initial evidence that 2D slice-based texture analysis can support early ASD classification and offer practical insights for optimizing neuroimaging-based diagnostic tools. Future work will explore deep learning approaches and multimodal integration to enhance robustness and clinical applicability.

The classification of autism spectrum disorder by machine learning methods on multiple datasets for four age groups

The presence of multiple dysmorphic features in some children with autism spectrum disorder (ASD) might identify distinct ASD phenotypes and serve as potential markers for understanding causes and prognoses. To evaluate dysmorphology in ASD, children aged 3-6years with ASD and non-ASD population controls (POP) from the Study to Explore Early Development were evaluated using a novel, systematic dysmorphology review approach. Separate analyses were conducted for non-Hispanic White, non-Hispanic Black, and Hispanic children. In each racial/ethnic group, ~ 17% of ASD cases were Dysmorphic compared with ~ 5% of POP controls. The ASD-POP differential was not explained by known genetic disorders or birth defects. In future epidemiologic studies, subgrouping ASD cases as Dysmorphic vs. Non-dysmorphic might help delineate risk factors for ASD.