Autism Diagnostic Interview-Revised Research Articles

A comprehensive review of neuroimaging-based AI models for autism classification

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by challenges in social interaction, communication, and behavior. Traditional diagnostic methods rely heavily on behavioral assessments, such as the Autism Diagnostic Observation Schedule (ADOS) and Autism Diagnostic Interview-Revised (ADI-R), which are often subjective and limited by the absence of definitive biomarkers. Recent advances in neuroimaging techniques, including structural MRI (sMRI), functional MRI (fMRI), and positron emission tomography (PET), have provided valuable insights into the structural and functional abnormalities associated with ASD. However, the analysis of neuroimaging data is complex, requiring sophisticated machine learning (ML) and artificial intelligence (AI) models. This paper reviews state-of-the-art machine learning approaches, such as Random Forests, Support Vector Machines (SVM), and deep learning models, aimed at improving the diagnostic accuracy of ASD. We particularly focus on multimodal deep learning models that integrate neuroimaging data with behavioral assessments. In this review, we critically evaluate the strengths and limitations of these approaches, and how emerging AI tools not only enhance classification accuracy but also hold potential for identifying biomarkers and tracking the disorder's progression. Our analysis underscores the need for advanced methods that exploit both spatial and temporal features of brain data to better understand ASD, paving the way for improved diagnostics and personalized treatments.

Use of Telehealth in Autism Spectrum Disorder Assessment in Children: Evaluation of anOnline Diagnostic Protocol Includingthe Brief Observation of Symptoms of Autism.

The COVID-19 pandemic revealed the need to develop the field of remote assessment for autism spectrum disorders (ASD). The purpose of the study was to evaluate an online assessment protocol that includes the Brief Observation of Symptoms of Autism (BOSA). The online protocol consisting of BOSA and the Autism Diagnostic Interview-Revised (ADI-R) was administered by experienced examiners to 29 children with suspected ASD. The participants were then evaluated by clinical psychologists in a standard clinical setting using the Autism Diagnostic Observation Schedule-2 (ADOS-2) and other methods, and the ASD diagnosis was confirmed or ruled out. The results show substantial to moderate inter-rater agreement between the online and face-to-face raters with the value of Cohen's Kappa = 0.66 (P < 0.001); this corresponds with 79.8% agreement. The sensitivity of the protocol was approx. 94.7%, the specificity was 70%, the positive predictive value was 85.7%, and the negative predictive value was 87.5%. Further, direct false positive or false negative diagnostic conclusions based on the online protocol were absent when the possible conclusion of "I cannot decide" was included. The items B9 Showing, B10 Spontaneous Initiation of Joint Attention, B1 Unusual Eye Contact, B3 Facial Expressions Directed to Others, and C2 Imagination/Creativity were shown to be well observable in BOSA when related to ADOS-2 scoring. The results indicate that the protocol consisting of BOSA and ADI-R administered by an experienced examiner is a promising combination of tools for remote autism assessment.

Structure of Restricted Repetitive Behaviors of Individuals Referred for Autism Spectrum Disorder Assessment.

Restricted and/or repetitive displays of behavior, interests, or activities (RRBs) are one of the core symptom domains of autism spectrum disorder (ASD). Current and past research indicates two 'clusters' of RRBs in children with ASD: repetitive sensorimotor (e.g., hand/finger and more complex motor mannerisms) and insistence on sameness (e.g., resistance to changes in the environment) behaviors. The current study aims to fill a gap by examining how RRBs may diverge in individuals with ASD and with other neurodevelopmental disorders (ONDD) in a clinical sample. A total of 558 individuals were seen at a tertiary care clinic for a comprehensive clinical assessment of ASD. The sample was split into ASD (n = 292 individuals) and ONDD (n = 266) groups based on clinical diagnosis. Exploratory factor analyses were conducted using Autism Diagnostic Interview-Revised (ADI-R) RRB item scores for the overall sample, the ASD group, and the ONDD group. Exploratory factor analysis of ADI-R RRB items indicated a 2-factor solution for the full sample and ASD group. Items loaded onto two factors comprised of "Repetitive Sensorimotor" and "Insistence on Sameness" behaviors, consistent with previous literature. Results demonstrated a unique loading pattern for the non-ASD group, with items clustering into "Higher Order" (e.g., circumscribed interests) and "Lower Order" (e.g., hand and finger mannerisms) behaviors. The results of the current study may point towards using RRBs to guide screening of children who are referred for an ASD evaluation to better identify children who are at higher risk of having ASD.

Age-related differences in delta-beta phase-amplitude coupling in autistic individuals

ObjectiveWe aim to investigate the relationship between the core symptoms of autism, anxiety levels, and attention deficit hyperactivity disorder (ADHD) traits, and a non-autism-specific, neurophysiological metric, the Delta-Beta phase-amplitude coupling (PAC), extracted from the resting-state EEG for autistic and non-autistic populations across three different age groups (children, adolescents, and adults). MethodsWe analyze the eyes-open resting-state EEG of 371 individuals. We applied a phase de-biasing PAC algorithm expected to result in a more accurate PAC estimate than other PAC methodologies available in the literature. ResultsIn the adult group, we found a significant increase of the delta-beta PAC in the autistic subgroup who met the Autism Diagnostic Observation Schedule-2 (ADOS-2) Autism Diagnostic Interview-Revised (ADR-R) ADOS-2/ADI-R threshold compared to non-autistic individuals. The differences seem age-specific since we found no statistically significant differences in the children and adolescent populations. Moreover, we found a significant positive correlation with the restricted and repetitive behaviours score of the ADOS-2 diagnostic instrument and with ADHD hyperactivity/impulsivity in the entire autistic cohort. ConclusionsThe neurophysiological differences we found only in the autistic individuals that meet the thresholds also point out the need for future studies that look for autistic neurodiverse subgroups beyond age. SignificanceThe delta-beta debiasing PAC (dPAC) may potentially serve as a severity biomarker in the autistic population.

Dysregulation of miR-146a in human milk of mothers having children with autism.

Autism spectrum disorder (ASD) is a set of neurobehavioral manifestations that impose poor social interaction and stereotyped repetitive patterns. Several mircoRNA (miRNA) dysregulations underpin ASD pathophysiology via impairing the neurogenic niches. For instance, miR-146a and miR-106 differential expressions are linked to deregulation of ASD-related genes and the severity of clinical symptoms, respectively. Breastfeeding provides newborns with many bioactive compounds that support their neurodevelopment including miRNA. Our pilot study evaluated the expression pattern of miR-106a and miR-146a in human milk (HM) of nursing mothers (n = 36) having autistic children compared to age-matched counterparts (n = 36) with neurotypical children as controls. Under sterile conditions, breast milk samples were collected using manual sucking pumps and centrifuged to separate the fat layer. Total RNA was extracted from the lipid fraction, and the expression profiles of both miR-106a and miR-146a were evaluated using quantitative real-time polymerase chain reaction. Among the test group, we reported some factors that were previously linked to HM miRNA perturbations: gestational diabetes, hypertension, and cesarean delivery. HM miR-106a showed comparable expression levels in both mother groups (p = 0.8681), whereas HM miR-146a was significantly downregulated in mothers with autistic children compared to controls (p = 0.0399). Alternatively, HM miR-106 levels were positively associated with two ASD clinical parameters: Childhood Autism Rating Scale (CARS) and communication and language domain of Autism Diagnostic Interview-Revised (ADI-R) (r = 0.6452, p = 0.0003 and r = 0.3958, p = 0.0410, respectively). The receiver operating characteristic (ROC) curves of both maternal HM miR-106a and miR-146a showed poor fitness as predictive biomarkers for ASD. Our findings suggest that the miR-146a differential expression in ASD children may originate at infancy during the lactation period. Thus, maternal pre- and postnatal health care is critical to maintain optimal miRNome in breast milk.

Assessment of Psychotic Risk in a Sample of Children and Adolescents with Autism Spectrum Disorder Compared to a Group of "Clinical High Risk" Patients: A Preliminary Study.

Autism spectrum disorder and psychotic risk show several overlapping symptoms, so differential diagnosis is often difficult. In addition, there is a high rate of comorbidity between the two conditions, which further complicates the work of clinicians. We evaluated the presence of subthreshold psychotic symptoms and/or defined psychotic risk syndromes in autistic children and adolescents; we compared the prevalence, type, and severity of psychotic risk symptoms with those of a group of non-autistic patients at clinical high risk for psychosis (CHR-P). 2. In total, 23 autistic patients and 14 CHR-P patients without autism (aged 8-17) were enrolled in the study. The main assessment was made through clinical interviews for autism (Autism Diagnostic Observation Schedule, Second Edition-ADOS-2, Autism Diagnostic Interview, Revised-ADI-R) and psychotic risk (Schizophrenia Proneness Instrument, Child and Youth version-SPI-CY, Structured Interview for Psychosis Risk Syndromes-SIPS). 3. No above-threshold psychotic risk symptoms were detected in our autistic patients, but subthreshold psychotic symptoms were identified in all areas. Specific items from all four dimensions of SIPS appear to be more specific for psychotic risk than autism without comorbidity. 4. An a priori screening of psychotic risk in neurodiverse populations is fundamental to prevent more severe conditions. Research should clarify the effective specificity of the available tools to modify them to improve their detection capability.

A Preliminary Trial of an Early Surveillance Program for Autism and Developmental Delays within General Practices

AbstractThere are inequities in engagement with established early childhood developmental surveillance programs, eclipsing disadvantaged families. The current study sought to address this by dovetailing developmental surveillance with immunization visits and other opportunistic contacts with children at general practices). While 53 General Practices were recruited, significant COVID-19 disruptions resulted in only 81 children being screened (both parent-administered and GP completed). Of the 81 children, 11 screened positive and all of them along with 5% of screen negatives (i.e. 4 children) received clinician-administered reference-standard assessment for autism and developmental delay (DD) using Autism Diagnostic Observation Schedule (ADOS), Autism Diagnostic Interview Schedule –Revised (ADI-R), and Mullen Scales of Early Learning (MSEL). All children found by reference-standard assessment to have probable autism and/or DD had screened positive during the screening process, and 90.9% of children who screened positive were found by reference-standard assessment to have probable DD or autism. The findings provide early evidence for the feasibility and usefulness of parent completed and GP administered developmental measures during opportunistic contacts with GPs as a promising method to facilitate early identification of DD or autism.

Experiences and concerns of parents of children with a 16p11.2 deletion or duplication diagnosis: a reflexive thematic analysis

Background16p11.2 proximal deletion and duplication syndromes (Break points 4–5) (593KB, Chr16; 29.6-30.2mb - HG38) are observed to have highly varied phenotypes, with a known propensity for lifelong psychiatric problems. This study aimed to contribute to a research gap by qualitatively exploring the challenges families with 16p11.2 deletion and duplication face by answering three research questions: (1) What are parents’ perceptions of the ongoing support needs of families with children who have 16p11.2 living in the UK?; (2) What are their experiences in trying to access support?; (3) In these regards, do the experiences of parents of children with duplication converge or vary from those of parents of children with 16p11.2 deletion?Methods33 parents with children (aged 7–17 years) with 16p11.2 deletion or duplication participated in structured interviews, including the Autism Diagnostic Interview– Revised (ADI-R). Their answers to the ADI-R question ‘what are your current concerns’ were transcribed and subsequently analysed using Braun and Clarke’s six step reflexive thematic analysis framework.ResultsThree themes were identified: (1) Child is Behind Peers (subthemes: developmentally; academically; socially; emotionally); (2) Metabolism and Eating Patterns and; (3) Support (subthemes: insufficient support available; parent has to fight to access support; COVID-19 was a barrier to accessing support; 16p11.2 diagnosis can be a barrier to support, child is well-supported).ConclusionsParents of children with either 16p11.2 deletion or duplication shared similar experiences. However, metabolism concerns were specific to parents of children with 16p11.2 deletion. The theme Child is Behind Peers echoed concerns raised in previous Neurodevelopmental Copy Number Variant research. However, there were some key subthemes relating to research question (2) which were specific to this study. This included parents’ descriptions of diagnostic overshadowing and the impact of a lack of eponymous name and scant awareness of 16p11.2.

Language defi cit in boys with autism spectrum disorder in relation to maternal reproductive health, endocrine disruptors, and delivery method.

The presented research aimed to identify prenatal factors involved in abnormal neurodevelopment and postnatal manifestation of an autistic phenotype in 107 boys (average age 4.31 ± 2.24 years). Their biological mothers were asked to fill out a comprehensive questionnaire about their reproductive health, infections during pregnancy, oral contraceptive intake before conception, and potential substance abuse before and during pregnancy as well as delivery and newborn information. The boys were subsequently diagnosed with autism spectrum disorder (ASD) using the combination of Autism Diagnostic Observation Schedule (ADOS-2) and Autism Diagnostic Interview - Revised (ADI-R). Based on the ADOS-2 module chosen during diagnosis, boys diagnosed with Module 1 can be classified as nonverbal or minimally verbal (N = 68), while those diagnosed using Module 3 are fully verbal (N = 39). According to our results, reproductive health related to the length of the menstrual cycle before pregnancy with the autistic child seems to play a role with regards to the severity of the disorder (P = 0.017) as well as the number of previous pregnancies (P = 0.026). Mothers of nonverbal children reported to have had a much shorter menstrual cycle (27.35 ± 6.60 days) than those with verbal children (30.14 ± 4.44 days) and reported more previous pregnancies (0.93 ± 1.07 vs. 0.51 ± 0.91), while not reporting the number of live births before they had the autistic child. Children who were later diagnosed as non-verbal had a longer delivery time (from 2 to 48 hours; on average 11.13 hours, SD = 9.49) than verbal ones (between 1 and 27 hours, which was on average 7.09 hours, SD = 8.91), P = 0.0182. Delivery method didn't play a role in this context, and neither did the type of conception (natural, insemination, etc.). Studying the involvement of prenatal factors in the etiology of autism based on the speech of the child seems to be a promising approach.

Age at initial diagnosis of autism spectrum disorders: a retrospective comparison of screening techniques between the southern and northwestern regions of Switzerland.

Early diagnosis of autism spectrum disorders (ASD) offers the possibility of early intervention and, in turn, gains in adaptive behaviour, language and cognition. The aim of the present study was to analyse whether age at diagnosis of autism spectrum disorders decreased in two regions of Switzerland from 2006 to 2016 following the implementation of different screening and referral techniques. In southern Switzerland, systematic paediatric screening using the Modified Checklist for Autism (M-CHAT) in toddlers was implemented in 2013, whereas in northwestern Switzerland, periodic trainings were used to increase paediatrician awareness of ASD. We investigated which method was associated with a younger average age at diagnosis. We conducted a retrospective, two-centre study searching clinical records of children and adolescents (aged 0-16 years) diagnosed with ASD in two neuropaediatric departments at Swiss hospitals between January 2006 and December 2016. All patients were diagnosed via a standardised evaluation based on two approved diagnostic tests: the Autism Diagnostic Observation Schedule-Second Edition (ADOS-2) and the Autism Diagnostic Interview-Revised (ADI-R). In southern Switzerland, training and subsequent widespread use of the M-CHAT among paediatricians appeared to contribute to a significantly younger age at diagnosis. Age at diagnosis did not significantly decrease during the same period in northwestern Switzerland. Our results point to the possibility of successfully reducing age at diagnosis in specific geographic areas through the implementation of screening questionnaires, such as the M-CHAT, at year 2 well-baby visits.

Autism spectrum disorder prevalence in Italy: a nationwide study promoted by the Ministry of Health

BackgroundThis nationwide study aimed to estimate Autism Spectrum Disorder (ASD) prevalence in 7–9-year-old Italian children. Promoted by Italy's Ministry of Health and coordinated by the National Observatory for Autism at the National Institute of Health, it covered schools in northern (Lecco and Monza-Brianza), central (Rome and its province), and southern (Palermo and its province) regions from February 24, 2016, to February 23, 2018, using a multi-stage approach defined by the European Union's ASD network.MethodsPhase one identified ASD-diagnosed children in mainstream schools through local Ministry of Education (MoE) disability registries. Phase two had a subset of schools screen 7–9-year-olds using the Social Communication Questionnaire-Life version (SCQ-L). Those with SCQ-L scores of 15 + underwent clinical consultation for ASD symptoms, cognitive abilities, and life skills. To counter potential false negatives, 20% scoring 11–14 were randomly assessed via Autism Diagnostic Interview-Revised (ADI-R).ResultsMoE data revealed 9.8 per 1000 certified ASD children in the north, 12.2 in the central, and 10.3 in the south. In phase two, 35,823 SCQ-L questionnaires were distributed across 198 schools (northern: 11,190 in 49 schools, central: 13,628 in 87 schools, southern: 11,005 in 62 schools). Of SCQ-L respondents, 2.4% (n = 390) scored above the 15 cutoff. Among these, 100 had ASD diagnoses, and 50 had other diagnoses. Among 115 families assessed, 16.5% (n = 19) received ASD diagnoses.ConclusionsThe estimated prevalence of ASD in Italy was 13.4 (11.3–16.0) per 1,000 children aged 7–9 years, with a male-to-female ratio of 4.4:1. It will guide national policies in enhancing services tailored to the specific needs of autistic children.

0789 Sleep Spindle Density is Associated with Core Symptom Severity in Autism Spectrum Disorder

Abstract Introduction Converging evidence suggests that dysregulation of thalamocortical interactions, mediated by the thalamic reticular nucleus (TRN), contributes to the expression of autism spectrum disorder (ASD). Risk genes for ASD have been found to affect the TRN early in development. Two hallmarks of NREM sleep –slow oscillations (SO) and sleep spindles– reflect TRN function and distinct thalamic/thalamocortical circuits. While studies demonstrate disrupted sleep quality that correlates with symptoms in children with ASD, studies of spindles in ASD have produced inconsistent findings. It is yet unclear if and how spindle properties are associated with ASD symptoms. Herein, based on studies implicating TRN-mediated thalamocortical dysregulation in ASD, we hypothesized that 1) sleep spindles are reduced in ASD; 2) spindle abnormalities correlate with worse clinical symptoms and cognition. Methods Forty participants with ASD (32 males, age: 11.8 ± 3.2; range 6-15) and 46 typically developing (TD) controls (32 males, age: 11.2 ± 2.3; range 6-15) participated in the study. All participants underwent ambulatory overnight polysomnography (PSG) for sleep assessment after systematic desensitization. The Autism Diagnostic Observation Schedule (ADOS) and the Autism Diagnostic Interview-Revised (ADI-R) were used to assess core behavioral symptoms. Results Spindle density was significantly lower in ASD compared to TD participants (ASD: 2.5 ± 2.0; TD: 3.0 ± 2.1; d= 0.45, p=0.01). In the ASD group, less spindle density was associated with more severe behavioral symptoms, as measured by the total ADOS score (R2=0.15, p=0.01) and the three main subscales of the ADI-R (reciprocal social interaction: R2=0.22, p=0.002; communication and language: R2=0.19, p=0.003; and restricted and repetitive, stereotyped interests and behaviors: R2=0.14, p=0.01). Conclusion Alterations in the maturational trajectory of spindles are related to the core symptoms of ASD individuals. The results have the potential to identify novel biomarkers that index distinct aspects of ASD pathophysiology and point to multiple neural mechanisms underlying disease heterogeneity. The study may contribute to producing novel evidence to elucidate the bidirectional relationship between sleep and ASD. Support (if any) Simons Foundation Autism Research Initiative grant (177986).

Stratifying ASD and characterizing the functional connectivity of subtypes in resting-state fMRI

Although stratifying autism spectrum disorder (ASD) into different subtypes is a common effort in the research field, few papers have characterized the functional connectivity alterations of ASD subgroups classified by their clinical presentations. This is a case-control rs-fMRI study, based on large samples of open database (Autism Brain Imaging Data Exchange, ABIDE). The rs-MRI data from n=415 ASD patients (males n=357), and n=574 typical development (TD) controls (males n=410) were included. Clinical features of ASD were extracted and classified using data from each patient's Autism Diagnostic Interview-Revised (ADI-R) evaluation. Each subtype of ASD was characterized by local functional connectivity using regional homogeneity (ReHo) for assessment, remote functional connectivity using voxel-mirrored homotopic connectivity (VMHC) for assessment, the whole-brain functional connectivity, and graph theoretical features. These identified imaging properties from each subtype were integrated to create a machine learning model for classifying ASD patients into the subtypes based on their rs-fMRI data, and an independent dataset was used to validate the model. All ASD participants were classified into Cluster-1 (patients with more severe impairment) and Cluster-2 (patients with moderate impairment) according to the dimensional scores of ADI-R. When compared to the TD group, Cluster-1 demonstrated increased local connection and decreased remote connectivity, and widespread hyper- and hypo-connectivity variations in the whole-brain functional connectivity. Cluster-2 was quite similar to the TD group in both local and remote connectivity. But at the level of whole-brain functional connectivity, the MCC-related connections were specifically impaired in Cluster-2. These properties of functional connectivity were fused to build a machine learning model, which achieved ~75% for identifying ASD subtypes (Cluster-1 accuracy = 81.75%; Cluster-2 accuracy = 76.48%). The stratification of ASD by clinical presentations can help to minimize disease heterogeneity and highlight the distinguished properties of brain connectivity in ASD subtypes.

The Neurogenetics of Functional Connectivity Alterations in Autism: Insights From Subtyping in 657 Individuals

BackgroundThere is little consensus and controversial evidence on anatomical alterations in the brains of people with autism spectrum disorder (ASD), due in part to the large heterogeneity present in ASD, which in turn is a major drawback for developing therapies. One strategy to characterize this heterogeneity in ASD is to cluster large-scale functional brain connectivity profiles. MethodsA subtyping approach based on consensus clustering of functional brain connectivity patterns was applied to a population of 657 autistic individuals with quality-assured neuroimaging data. We then used high-resolution gene transcriptomic data to characterize the molecular mechanism behind each subtype by performing enrichment analysis of the set of genes showing a high spatial similarity with the profiles of functional connectivity alterations between each subtype and a group of typically developing control participants. ResultsTwo major stable subtypes were found: subtype 1 exhibited hypoconnectivity (less average connectivity than typically developing control participants) and subtype 2, hyperconnectivity. The 2 subtypes did not differ in structural imaging metrics in any of the analyzed regions (68 cortical and 14 subcortical) or in any of the behavioral scores (including IQ, Autism Diagnostic Interview, and Autism Diagnostic Observation Schedule). Finally, only subtype 2, comprising about 43% of ASD participants, led to significant enrichments after multiple testing corrections. Notably, the dominant enrichment corresponded to excitation/inhibition imbalance, a leading well-known primary mechanism in the pathophysiology of ASD. ConclusionsOur results support a link between excitation/inhibition imbalance and functional connectivity alterations, but only in one ASD subtype, overall characterized by brain hyperconnectivity and major alterations in somatomotor and default mode networks.

Adolescent with 22q11.2 Microdeletion Syndrome, Cognitive Delay, Autism Spectrum Disorder and Psychotic Features

Microdeletion 22q11.2 syndrome (22q11.2DS) is a common microdeletion syndrome, also described as DiGeorge syndrome (DGS). It has a prevalence estimated to be within 1 per 2148 livebirths. Eighty-four percent of the children with 22q11DS had at least one psychiatric disorder, including autism spectrum disorder (ASD), schizophrenia, neurocognitive delay and other neuropsychiatric disorders. Sometimes is not easy to detect psychiatric diagnosis in developmental disorders, so it is important to look for psychiatric symptoms, etiological factors and clinical report about child development, including parent’s perceptions. We present a female adolescent patient with a mild cognitive delay, and underdiagnosed ASD, that had a microdeletion 22q11.2 Syndrome, identified at the first year of life. She was evaluated in psychiatric consultation only at 15 years, for psychotic symptomatology. At that time some ASD features were identified, and lately confirmed, with parental information and psychological evaluation instruments, like Autism Diagnostic Interview Revised (ADI-R). This paper aims to alert to the possibility of confluence of 22q11.2DS with ASD and psychotic symptoms, at the same time. We also want to enhance the importance of a multidisciplinary team in developmental disorders, paying attention to developmental report and parents ‘information, which may possibility early intervention and an accurate diagnosis.

Web-based autism screening using facial images and convolutional neural network

Developmental disabilities such as autism spectrum disorder (ASD) affect a person’s ability to interact socially, and communicate effectively and also cause behavioral issues. Children with ASD cannot be cured but they might benefit from early intervention to enhance their cognitive abilities, favorite their growth , and affect their lives and families in a positive way. Multiple standard ASD screening tools are used such as the autism diagnostic observational schedule (ADOS) and the autism diagnostic interview (ADI), which are known to be lengthy and challenging without specialist training to administrate and score. The process of ASD assessment can be time-consuming and costly, and the growing number of autistic cases worldwide indicates an urgent need for a quick, simple, and dependable self-administered autism screening tool that may be used if a child displays some of the common signs of autism, and to ensure whether or not he should seek professional full ASD diagnosis. According to a number of studies, ASD individuals exhibit facial phenotypes that are distinct from those of normally developing children. Furthermore, convolutional neural networks (CNN) have mostly found utility in image classification applications due to their high classification accuracy. Using facial images, a dense convolutional network (Densenet) model, and cloud-based advantages, in this paper we proposed a practical, fast, and easy-to-use ASD online screening approach. Easily available through the internet via the link “https://asd-detector.herokuapp.com/”, our suggested web-based screening instrument may be a practical and trustworthy tool for practitioners in their ASD diagnostic procedures with a 98 percent testing dataset classification accuracy.

Psychometric properties of the Mandarin version of the autism diagnostic observation Schedule-Generic

The diagnosis of autism spectrum disorder (ASD), involving multiple components of clinical assessments, is challenging. The Autism Diagnostic Observation Schedule-Generic (ADOS-G), one of the standardized and validated instruments for ASD diagnostic evaluation, has been widely used in many countries. With the preparation of the Mandarin version of the ADOS-G (Mandarin-ADOS-G), this study aims to examine its psychometric properties, including reliability and validity. The sample included 554 individuals clinically diagnosed with ASD (477 males, 86.1%) and 50 typically developing (TD) individuals (29 males, 58.0%) who were assessed with different modules of the Mandarin-ADOS-G between 4.1 and 34.0 years old with a mean age of 13.0 years (Module 1, n=40; Module 2, n=46; Module 3, n=275; Module 4, n=243). We evaluated the inter-rater reliability, test-retest reliability, internal consistency, and concurrent validity with the Chinese Autism Diagnostic Interview-Revised (ADI-R) and Social Responsiveness Scale (SRS) caregiver-report and self-report forms. The discriminative validity of Mandarin-ADOS-G was also examined. The Mandarin-ADOS-G demonstrated good inter-rater reliability (agreement of ADOS classification 0.91), good test-retest reliability (intraclass correlations 0.55-0.73), and low to high good internal consistency (Cronbach's alpha 0.27-0.86). The concurrent validity showed significant correlations with ADI-R (Pearson correlations 0.22-0.37) and the SRS caregiver-report form (Pearson correlations 0.15-0.23). Moreover, all Mandarin-ADOS-G domains successfully differentiated autistic individuals from TD individuals (all p-values <0.001). The Mandarin-ADOS-G is a reliable and valid instrument for assisting the diagnosis of ASD in the Mandarin-speaking population.

Gray matter asymmetry atypical patterns in subgrouping minors with autism based on core symptoms.

Abnormal gray matter (GM) asymmetry has been verified in autism spectrum disorder (ASD), which is characterized by high heterogeneity. ASD is distinguished by three core symptom domains. Previous neuroimaging studies have offered support for divergent neural substrates of different core symptom domains in ASD. However, no previous study has explored GM asymmetry alterations underlying different core symptom domains. This study sought to clarify atypical GM asymmetry patterns underlying three core symptom domains in ASD with a large sample of 230 minors with ASD (ages 7-18 years) and 274 matched TD controls from the Autism Brain Imaging Data Exchange I (ABIDE I) repository. To this end, the scores of the revised autism diagnostic interview (ADI-R) subscales were normalized for grouping ASD into three core-symptom-defined subgroups: social interaction (SI), verbal communication (VA), and restricted repetitive behaviors (RRB). We investigated core-symptom-related GM asymmetry alterations in ASD resulting from advanced voxel-based morphometry (VBM) by general linear models. We also examined the relationship between GM asymmetry and age and between GM asymmetry and symptom severity assessed by the Autism Diagnostic Observation Schedule (ADOS). We found unique GM asymmetry alterations underlying three core-symptom-defined subgroups in ASD: more rightward asymmetry in the thalamus for SI, less rightward asymmetry in the superior temporal gyrus, anterior cingulate and caudate for VA, and less rightward asymmetry in the middle and inferior frontal gyrus for RRB. Furthermore, the asymmetry indexes in the thalamus were negatively associated with ADOS_SOCIAL scores in the general ASD group. We also showed significant correlations between GM asymmetry and age in ASD and TD individuals. Our results support the theory that each core symptom domain of ASD may have independent etiological and neurobiological underpinnings, which is essential for the interpretation of heterogeneity and the future diagnosis and treatment of ASD.

Excitatory/inhibitory imbalance in autism: the role of glutamate and GABA gene-sets in symptoms and cortical brain structure

The excitatory/inhibitory (E/I) imbalance hypothesis posits that imbalance between excitatory (glutamatergic) and inhibitory (GABAergic) mechanisms underlies the behavioral characteristics of autism. However, how E/I imbalance arises and how it may differ across autism symptomatology and brain regions is not well understood. We used innovative analysis methods—combining competitive gene-set analysis and gene-expression profiles in relation to cortical thickness (CT) to investigate relationships between genetic variance, brain structure and autism symptomatology of participants from the AIMS-2-TRIALS LEAP cohort (autism = 359, male/female = 258/101; neurotypical control participants = 279, male/female = 178/101) aged 6–30 years. Using competitive gene-set analyses, we investigated whether aggregated genetic variation in glutamate and GABA gene-sets could be associated with behavioral measures of autism symptoms and brain structural variation. Further, using the same gene-sets, we corelated expression profiles throughout the cortex with differences in CT between autistic and neurotypical control participants, as well as in separate sensory subgroups. The glutamate gene-set was associated with all autism symptom severity scores on the Autism Diagnostic Observation Schedule-2 (ADOS-2) and the Autism Diagnostic Interview-Revised (ADI-R) within the autistic group. In adolescents and adults, brain regions with greater gene-expression of glutamate and GABA genes showed greater differences in CT between autistic and neurotypical control participants although in opposing directions. Additionally, the gene expression profiles were associated with CT profiles in separate sensory subgroups. Our results suggest complex relationships between E/I related genetics and autism symptom profiles as well as brain structure alterations, where there may be differential roles for glutamate and GABA.

Review of Progress in Diagnostic Studies of Autism Spectrum Disorder Using Neuroimaging.

This review article summarizes the recent advances in the diagnostic studies of autism spectrum disorders (ASDs) considering some of the most influential research articles from the last two decades. ASD is a heterogeneous neurodevelopmental disorder characterized by abnormalities in social interaction, communication, and behavioral patterns as well as some unique strengths and differences. The current diagnosis systems are based on autism diagnostic observation schedule (ADOS) or autism diagnostic interview-revised (ADI-R), but biological markers are also important for an effective diagnosis of ASDs. The amalgamation of neuroimaging techniques, such as structural and functional magnetic resonance imaging (sMRI and fMRI), with machine-learning and deep-learning approaches helps throw new light on typical biological markers of ASDs at the early stage of life. To assess the performance of a deep neural network, we develop a light-weighted CNN model for ASD classification. The overall accuracy, precision, and F1-score of the proposed model are 99.92%, 99.93% and 99.92%, respectively. All the neuroimaging studies we have reviewed can be divided into 3 categories, viz. thickness, volume and functional connectivity-based studies. We conclude with a discussion of the major findings of considered studies and promising directions for future research in this field.