Language and Speech Characteristics in Autism

Clinical Decision Support Using Speech Signal Analysis: Systematic Scoping Review of Neurological Disorders

BACKGROUND

Digital biomarkers are increasingly used in clinical decision support for various health conditions. Speech features as digital biomarkers can offer insights into underlying physiological processes due to the complexity of speech production. This process involves respiration, phonation, articulation, and resonance, all of which rely on specific motor systems for the preparation and execution of speech. Deficits in any of these systems can cause changes in speech signal patterns. Increasing efforts are being made to develop speech-based clinical decision support systems.

OBJECTIVE

This systematic scoping review investigated the technological revolution and recent digital clinical speech signal analysis trends to understand the key concepts and research processes from clinical and technical perspectives.

METHODS

A systematic scoping review was undertaken in 6 databases guided by a set of research questions. Articles that focused on speech signal analysis for clinical decision-making were identified, and the included studies were analyzed quantitatively. A narrower scope of studies investigating neurological diseases were analyzed using qualitative content analysis.

RESULTS

A total of 389 articles met the initial eligibility criteria, of which 72 (18.5%) that focused on neurological diseases were included in the qualitative analysis. In the included studies, Parkinson disease, Alzheimer disease, and cognitive disorders were the most frequently investigated conditions. The literature explored the potential of speech feature analysis in diagnosis, differentiating between, assessing the severity and monitoring the treatment of neurological conditions. The common speech tasks used were sustained phonations, diadochokinetic tasks, reading tasks, activity-based tasks, picture descriptions, and prompted speech tasks. From these tasks, conventional speech features (such as fundamental frequency, jitter, and shimmer), advanced digital signal processing-based speech features (such as wavelet transformation-based features), and spectrograms in the form of audio images were analyzed. Traditional machine learning and deep learning approaches were used to build predictive models, whereas statistical analysis assessed variable relationships and reliability of speech features. Model evaluations primarily focused on analytical validations. A significant research gap was identified: the need for a structured research process to guide studies toward potential technological intervention in clinical settings. To address this, a research framework was proposed that adapts a design science research methodology to guide research studies systematically.

CONCLUSIONS

The findings highlight how data science techniques can enhance speech signal analysis to support clinical decision-making. By combining knowledge from clinical practice, speech science, and data science within a structured research framework, future research may achieve greater clinical relevance.

Preliminary Results of Sensorimotor Room Training for the Improvement of Sensory and Motor Skills in Children with Autism Spectrum Disorders

Background: Motor skills in early and middle childhood are essential for physical play, social interactions, and academic development. Children with autism spectrum disorder (ASD) often exhibit atypical sensory responses, which can impact self-care and other developmental areas. This study explores the impact of sensory and motor rehabilitation using a Motor Sensory Room to stimulate motor development in children with ASD. Methods: Twenty-five children with ASD, preschool and school-aged (2-10 years), were divided into three groups based on the DSM-5 severity levels. The PEP-3 scale was used to assess cognitive, language, motor, emotional, social, and behavioral development. Ten children underwent a 3-month Motor Sensory Room intervention, and data were collected longitudinally. A control group of ten children, matched in age, sex, and diagnosis, did not receive the intervention. A 12-month follow-up is planned for all participants. Results: Children exhibited diverse profiles. Type B subjects displayed more severe symptoms, while Type A showed milder symptoms with better language and interpersonal skills. After the 3-month intervention, improvements were noted in several PEP-3 areas. For sensory levels (hypo-reactivity), the percentage of individuals at medium levels increased from 44% to 50%. For hyper-reactivity, the percentage at medium levels rose from 30% to 40%. Motor skills improved, with the percentage of individuals with high motor abilities rising from 20% to 25%. Relational behaviors also saw gains, with an increase from 50% to 55% in medium-level behaviors. The experimental group demonstrated better outcomes compared to the control group, particularly in sensory and motor skills. Conclusions: Preliminary findings suggest that sensory and motor training in a Motor Sensory Room improves sensory integration, motor coordination, and social interaction in children with ASD. Further research is needed to confirm long-term benefits.

Effectiveness of Equine-Assisted Activities and Therapies for Children with Autism Spectrum Disorder: An Update

BACKGROUND/OBJECTIVES

Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition requiring personalised therapeutic approaches. This study evaluated the effectiveness of Equine-Assisted Activities and Therapies (EAATs) in 86 children with varying ASD severity levels (levels 1-3).

METHODS

Vineland Adaptive Behaviour Scales and the Parenting Stress Index were used. Between May 2022 and October 2023, participants completed 20 weekly sessions of 45 min each, tailored to their individual needs.

RESULTS

Children with level 3 ASD demonstrated greater challenges in communication (level 1: 67.1 ± 29.0 vs. level 3: 30.0 ± 12.6; p < 0.001), daily living skills (81.0 ± 26.8 vs. 42.6 ± 18.1; p < 0.001), and socialisation (72.2 ± 23.2 vs. 37.3 ± 14.2; p < 0.001). Parental distress was higher in cases of greater ASD severity. Nevertheless, significant improvements were observed across the entire cohort in daily living skills (58.3 ± 25.5 vs. 67.8 ± 29.0; p = 0.023), with particularly notable outcomes in children with level 1 ASD (65.7 ± 26.9 vs. 81.0 ± 26.8; p = 0.010). While increases in socialisation were noted among children with level 1 ASD, these were not statistically significant (p = 0.073).

CONCLUSIONS

EAAT fosters improvements in daily living skills, particularly in children with level 1 ASD, and has a positive impact on socialisation. For children with more severe ASD, targeted interventions are required.

The use of formal language as a strong sign of verbal autistic children in diglossic communities: The case of Arabic

The current study aimed to investigate whether the use of formal language (Modern Standard Arabic [MSA]) by young children in diglossic Arab communities offers diagnostic insights, especially for verbal autistic children and to further explore this phenomenon. We used a cohort study design, with 4-6-year-old fluent first language Arabic-speaking children attending Arabic Kindergartens in two representative Kuwait governates. Reported cases for MSA use were assessed via a computer-based structured language test and corroborated cases were further assessed for exposure to sources of MSA, verbal IQ, temperamental characteristics, and autism spectrum disorder (ASD). Four children from the same class without developmental difficulties were selected for each MSA user as control group. The frequency of MSA use among verbal pre-schoolers was 0.46%. Use of MSA did not correlate with parents' education, amount of exposure to MSA, verbal IQ, but with severity of ASD. Predicted probability of ASD in the presence of MSA was 0.86. Executive functions of ASD-MSA users were similar to those of the control group and significantly higher than unselected autistic peers in the literature. The use of MSA has the potential to serve as a strong sign for the diagnosis of verbal autistic children, often missed or delayed in being diagnosed. We also discuss strategies via which language is acquired in ASD.

Traits of Developmental Disorders in Adults With Listening Difficulties Without Diagnosis of Autism Spectrum Disorder And/or Attention-Deficit/Hyperactivity Disorder

Background: Some individuals have a normal audiogram but have listening difficulties (LiD). As many studies have investigated the relationship between listening and developmental disorders, the traits of developmental disorders might explain the symptoms of LiD. In this study, we examined the traits of developmental disorders of adults with LiD to help clarify the cause of LiD symptoms. Methods: In total, 60 adults with LiD and 57 adults without LiD were included. Participants completed a questionnaire for the autism spectrum quotient (AQ) test, the Adult Attention-Deficit Hyperactivity Disorder Self-Rating Scale (A-ADHD), the Adolescent/Adult Sensory Profile (SP), and the severity of subjective LiD in daily life. Results: Before analysis, we excluded participants with LiD who were already diagnosed or met the criteria for autism spectrum disorder (ASD) or ADHD, and the results of the remaining 30 participants (50.0%) with LiD were analyzed. Adults with LiD showed higher scores than those without LiD in the AQ. Attention switching in the AQ and attention ability in the A-ADHD scale were correlated with the severity of LiD symptoms in everyday life. The AQ scores were also significantly correlated with subscales of the SP. Conclusions: Adults with LiD showed greater autistic traits than those without LiD; therefore, LiD symptoms are possibly related to autistic symptoms. Furthermore, adults with LiD might have attention disorder traits of both ASD and ADHD and sensory processing problems. These findings suggest that the attention problems in adults with LiD noted in previous studies might be related to these traits of developmental disorders.

Psychometric Properties of the Chinese Version of the Gilliam Autism Rating Scale-Third Edition (CV-GARS-3)

Gilliam Autism Rating Scale-Third Edition (GARS-3) serves as an effective screening tool for autism spectrum disorder (ASD) that is based on the latest and authoritative diagnostic criteria, however, there is a deficiency in adaptive research in China. We aimed to revise the Chinese version of GARS-3 (CV-GARS-3) and evaluate its psychometric characteristics, providing a theoretical basis for the improvement of ASD screening tools in China. This study developed CV-GARS-3 through translation and cultural adaptation of GARS-3. 362 ASD individuals, 126 typical development individuals, and 103 individuals with other disorders were recruited to analyze the psychometric characteristics of CV-GARS-3. The results showed that exploratory structural equation model demonstrated satisfactory goodness-of-fit. Within the non-verbal ASD samples, all items loaded on anticipated factors. Regarding verbal ASD samples, 3 items exhibited considerable cross-loadings and were categorized under unexpected factors. Meanwhile, acceptable criterion validity was reflected in the four subscales (r = 0.71) and the six subscales (r = 0.74). Satisfactory reliability was observed in the four subscales (Cronbach's α = 0.96, inter-rater consistency = 0.86, test-retest consistency = 0.87) and the six subscales (Cronbach's α = 0.94, inter-rater consistency = 0.81, test-retest consistency = 0.81). In addition, receiver operating characteristic analysis demonstrated that both the four subscales (sensitivity = 89%, specificity = 85%, accuracy = 88%) and the six subscales (sensitivity = 86%, specificity = 88%, accuracy = 86%) had outstanding screening effects. Therefore, the results suggested that the CV-GARS-3 is considered as a useful tool for the screening and auxiliary diagnosis of ASD. Notably, the expression of scale should be further improved to adapt the context of Chinese culture and achieve more precise diagnostic results.

Speech and language patterns in autism: Towards natural language processing as a research and clinical tool

Speech and language differences have long been described as important characteristics of autism spectrum disorder (ASD). Linguistic abnormalities range from prosodic differences in pitch, intensity, and rate of speech, to language idiosyncrasies and difficulties with pragmatics and reciprocal conversation. Heterogeneity of findings and a reliance on qualitative, subjective ratings, however, limit a full understanding of linguistic phenotypes in autism. This review summarizes evidence of both speech and language differences in ASD. We also describe recent advances in linguistic research, aided by automated methods and software like natural language processing (NLP) and speech analytic software. Such approaches allow for objective, quantitative measurement of speech and language patterns that may be more tractable and unbiased. Future research integrating both speech and language features and capturing "natural language" samples may yield a more comprehensive understanding of language differences in autism, offering potential implications for diagnosis, intervention, and research.

Decoding the genetic landscape of autism: A comprehensive review

BACKGROUND

Autism spectrum disorder (ASD) is a complex neurodevelopmental condition characterized by heterogeneous symptoms and genetic underpinnings. Recent advancements in genetic and epigenetic research have provided insights into the intricate mechanisms contributing to ASD, influencing both diagnosis and therapeutic strategies.

AIM

To explore the genetic architecture of ASD, elucidate mechanistic insights into genetic mutations, and examine gene-environment interactions.

METHODS

A comprehensive systematic review was conducted, integrating findings from studies on genetic variations, epigenetic mechanisms (such as DNA methylation and histone modifications), and emerging technologies [including Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas9 and single-cell RNA sequencing]. Relevant articles were identified through systematic searches of databases such as PubMed and Google Scholar.

RESULTS

Genetic studies have identified numerous risk genes and mutations associated with ASD, yet many cases remain unexplained by known factors, suggesting undiscovered genetic components. Mechanistic insights into how these genetic mutations impact neural development and brain connectivity are still evolving. Epigenetic modifications, particularly DNA methylation and non-coding RNAs, also play significant roles in ASD pathogenesis. Emerging technologies like CRISPR-Cas9 and advanced bioinformatics are advancing our understanding by enabling precise genetic editing and analysis of complex genomic data.

CONCLUSION

Continued research into the genetic and epigenetic underpinnings of ASD is crucial for developing personalized and effective treatments. Collaborative efforts integrating multidisciplinary expertise and international collaborations are essential to address the complexity of ASD and translate genetic discoveries into clinical practice. Addressing unresolved questions and ethical considerations surrounding genetic research will pave the way for improved diagnostic tools and targeted therapies, ultimately enhancing outcomes for individuals affected by ASD.

Diagnostic staging and stratification in psychiatry and oncology: clarifying their conceptual, epistemological and ethical implications

Staging and stratification are two diagnostic approaches that have introduced a more dynamic outlook on the development of diseases, thus participating in blurring the line between the normal and the pathological. First, diagnostic staging, aiming to capture how diseases evolve in time and/or space through identifiable and gradually more severe stages, may be said to lean on an underlying assumption of "temporal determinism". Stratification, on the other hand, allows for the identification of various prognostic or predictive subgroups based on specific markers, relying on a more "mechanistic" or "statistical" form of determinism. There are two medical fields in which these developments have played a significant role and have given rise to sometimes profound nosological transformations: oncology and psychiatry. Drawing on examples from these two fields, this paper aims to provide much needed conceptual clarifications on both staging and stratification in order to outline how several epistemological and ethical issues may, in turn, arise. We argue that diagnostic staging ought to be detached from the assumption of temporal determinism, though it should still play an essential role in adapting interventions to stage. In doing so, it would help counterbalance stratification's own epistemological and ethical shortcomings. In this sense, the reflections and propositions developed in psychiatry can offer invaluable insights regarding how adopting a more transdiagnostic and cross-cutting perspective on temporality and disease dynamics may help combine both staging and stratification in clinical practice.

A general principle governing neuronal evolution reveals a human-accelerated neuron type potentially underlying the high prevalence of autism in humans

The remarkable ability of a single genome sequence to encode a diverse collection of distinct cell types, including the thousands of cell types found in the mammalian brain, is a key characteristic of multicellular life. While it has been observed that some cell types are far more evolutionarily conserved than others, the factors driving these differences in evolutionary rate remain unknown. Here, we hypothesized that highly abundant neuronal cell types may be under greater selective constraint than rarer neuronal types, leading to variation in their rates of evolution. To test this, we leveraged recently published cross-species single-nucleus RNA-sequencing datasets from three distinct regions of the mammalian neocortex. We found a strikingly consistent relationship where more abundant neuronal subtypes show greater gene expression conservation between species, which replicated across three independent datasets covering >106 neurons from six species. Based on this principle, we discovered that the most abundant type of neocortical neurons-layer 2/3 intratelencephalic excitatory neurons-has evolved exceptionally quickly in the human lineage compared to other apes. Surprisingly, this accelerated evolution was accompanied by the dramatic down-regulation of autism-associated genes, which was likely driven by polygenic positive selection specific to the human lineage. In sum, we introduce a general principle governing neuronal evolution and suggest that the exceptionally high prevalence of autism in humans may be a direct result of natural selection for lower expression of a suite of genes that conferred a fitness benefit to our ancestors while also rendering an abundant class of neurons more sensitive to perturbation.

A cross-dataset study on automatic detection of autism spectrum disorder from text data

OBJECTIVE

The goals of this article are as follows. First, to investigate the possibility of detecting autism spectrum disorder (ASD) from text data using the latest generation of machine learning tools. Second, to compare model performance on two datasets of transcribed statements, collected using two different diagnostic tools. Third, to investigate the feasibility of knowledge transfer between models trained on both datasets and check if data augmentation can help alleviate the problem of a small number of observations.

METHOD

We explore two techniques to detect ASD. The first one is based on fine-tuning HerBERT, a BERT-based, monolingual deep transformer neural network. The second one uses the newest, multipurpose text embeddings from OpenAI and a classifier. We apply the methods to two separate datasets of transcribed statements, collected using two different diagnostic tools: thought, language, and communication (TLC) and autism diagnosis observation schedule-2 (ADOS-2). We conducted several cross-dataset experiments in both a zero-shot setting and a setting where models are pretrained on one dataset and then training continues on another to test the possibility of knowledge transfer.

RESULTS

Unlike previous studies, the models we tested obtained average results on ADOS-2 data but reached very good performance of the models in TLC. We did not observe any benefits from knowledge transfer between datasets. We observed relatively poor performance of models trained on augmented data and hypothesize that data augmentation by back translation obfuscates autism-specific signals.

CONCLUSION

The quality of machine learning models that detect ASD from text data is improving, but model results are dependent on the type of input data or diagnostic tool.

Assessing the Impact of Bilingualism on the Linguistic Skills of Children with Autism Spectrum Disorder (ASD) in Greece: A Scoping Review

(1) Background and Objectives: This review aims to identify the latest literature on the possible effect of bilingualism on the linguistic skills of children with autism spectrum disorder (ASD) residing in Greece. (2) Materials and Methods: The literature was searched in the databases of Scopus and PubMed by selecting articles and by reviewing four studies published in peer-reviewed journals. This Scoping Review is based on the standards of PRISMA recommendations for scoping reviews, while the PCC framework was used as a guide to construct clear and meaningful objectives and eligibility criteria. (3) Results: The publications included in the review addressed a variety of language-related skills, including morphology, the syntax-pragmatics interface, narrative ability, as well as both receptive and expressive language skills. (4) Conclusions: Three out of four studies provide evidence that bilingual ASD children are not disadvantaged compared to monolingual peers but rather enjoy some benefits, to a certain extent, due to bilingualism. However, the number of the reviewed studies as well as the limitations of the studies themselves render this conclusion tentative. Additionally, the findings set guidelines that speech therapists, educators, psychologists, and doctors in the Greek context need to follow when treating or educating bilingual children with ASD.

Neural correlates of audiovisual narrative speech perception in children and adults on the autism spectrum: A functional magnetic resonance imaging study

Autistic individuals show substantially reduced benefit from observing visual articulations during audiovisual speech perception, a multisensory integration deficit that is particularly relevant to social communication. This has mostly been studied using simple syllabic or word-level stimuli and it remains unclear how altered lower-level multisensory integration translates to the processing of more complex natural multisensory stimulus environments in autism. Here, functional neuroimaging was used to examine neural correlates of audiovisual gain (AV-gain) in 41 autistic individuals to those of 41 age-matched non-autistic controls when presented with a complex audiovisual narrative. Participants were presented with continuous narration of a story in auditory-alone, visual-alone, and both synchronous and asynchronous audiovisual speech conditions. We hypothesized that previously identified differences in audiovisual speech processing in autism would be characterized by activation differences in brain regions well known to be associated with audiovisual enhancement in neurotypicals. However, our results did not provide evidence for altered processing of auditory alone, visual alone, audiovisual conditions or AV- gain in regions associated with the respective task when comparing activation patterns between groups. Instead, we found that autistic individuals responded with higher activations in mostly frontal regions where the activation to the experimental conditions was below baseline (de-activations) in the control group. These frontal effects were observed in both unisensory and audiovisual conditions, suggesting that these altered activations were not specific to multisensory processing but reflective of more general mechanisms such as an altered disengagement of Default Mode Network processes during the observation of the language stimulus across conditions.

Language Delay in Patients with CLN2 Disease: Could It Support Earlier Diagnosis?

Neuronal ceroid lipofuscinosis type 2 (CLN2 disease) is a rare pediatric disorder associated with rapid neurodegeneration, and premature death in adolescence. An effective enzyme replacement therapy (cerliponase alfa) has been approved that can reduce this predictable neurological decline. The nonspecific early symptoms of CLN2 disease frequently delay diagnosis and appropriate management. Seizures are generally recognized as the first presenting symptom of CLN2 disease, but emerging data show that language delay may precede this. An improved understanding of language deficits in the earliest stage of CLN2 disease may support the early identification of patients. In this article, CLN2 disease experts examine how language development is affected by CLN2 disease in their clinical practices. The authors' experiences highlighted the timings of first words and first use of sentences, and language stagnation as key features of language deficits in CLN2 disease, and how deficits in language may be an earlier sign of the disease than seizures. Potential challenges in identifying early language deficits include assessing patients with other complex needs, and recognizing that a child's language abilities are not within normal parameters given the variability of language development in young children. CLN2 disease should be considered in children presenting with language delay and/or seizures to facilitate earlier diagnosis and access to treatment that can significantly reduce morbidity.

Connectome-based prediction of the severity of autism spectrum disorder

Background

Autism spectrum disorder (ASD) is characterized by social and behavioural deficits. Current diagnosis relies on behavioural criteria, but machine learning, particularly connectome-based predictive modelling (CPM), offers the potential to uncover neural biomarkers for ASD.

Objective

This study aims to predict the severity of ASD traits using CPM and explores differences among ASD subtypes, seeking to enhance diagnosis and understanding of ASD.

Methods

Resting-state functional magnetic resonance imaging data from 151 ASD patients were used in the model. CPM with leave-one-out cross-validation was conducted to identify intrinsic neural networks that predict Autism Diagnostic Observation Schedule (ADOS) scores. After the model was constructed, it was applied to independent samples to test its replicability (172 ASD patients) and specificity (36 healthy control participants). Furthermore, we examined the predictive model across different aspects of ASD and in subtypes of ASD to understand the potential mechanisms underlying the results.

Results

The CPM successfully identified negative networks that significantly predicted ADOS total scores [r (df = 150) = 0.19, P = 0.008 in all patients; r (df = 104) = 0.20, P = 0.040 in classic autism] and communication scores [r (df = 150) = 0.22, P = 0.010 in all patients; r (df = 104) = 0.21, P = 0.020 in classic autism]. These results were reproducible across independent databases. The networks were characterized by enhanced inter- and intranetwork connectivity associated with the occipital network (OCC), and the sensorimotor network (SMN) also played important roles.

Conclusions

A CPM based on whole-brain resting-state functional connectivity can predicted the severity of ASD. Large-scale networks, including the OCC and SMN, played important roles in the predictive model. These findings may provide new directions for the diagnosis and intervention of ASD, and maybe could be the targets in novel interventions.

Diverging asymmetry of intrinsic functional organization in autism

Autism is a neurodevelopmental condition involving atypical sensory-perceptual functions together with language and socio-cognitive deficits. Previous work has reported subtle alterations in the asymmetry of brain structure and reduced laterality of functional activation in individuals with autism relative to non-autistic individuals (NAI). However, whether functional asymmetries show altered intrinsic systematic organization in autism remains unclear. Here, we examined inter- and intra-hemispheric asymmetry of intrinsic functional gradients capturing connectome organization along three axes, stretching between sensory-default, somatomotor-visual, and default-multiple demand networks, to study system-level hemispheric imbalances in autism. We observed decreased leftward functional asymmetry of language network organization in individuals with autism, relative to NAI. Whereas language network asymmetry varied across age groups in NAI, this was not the case in autism, suggesting atypical functional laterality in autism may result from altered developmental trajectories. Finally, we observed that intra- but not inter-hemispheric features were predictive of the severity of autistic traits. Our findings illustrate how regional and patterned functional lateralization is altered in autism at the system level. Such differences may be rooted in atypical developmental trajectories of functional organization asymmetry in autism.

Personalized estimates of brain cortical structural variability in individuals with Autism spectrum disorder: the predictor of brain age and neurobiology relevance

BACKGROUND

Autism spectrum disorder (ASD) is a heritable condition related to brain development that affects a person's perception and socialization with others. Here, we examined variability in the brain morphology in ASD children and adolescent individuals at the level of brain cortical structural profiles and the level of each brain regional measure.

METHODS

We selected brain structural MRI data in 600 ASDs and 729 normal controls (NCs) from Autism Brain Imaging Data Exchange (ABIDE). The personalized estimate of similarity between gray matter volume (GMV) profiles of an individual to that of others in the same group was assessed by using the person-based similarity index (PBSI). Regional contributions to PBSI score were utilized for brain age gap estimation (BrainAGE) prediction model establishment, including support vector regression (SVR), relevance vector regression (RVR), and Gaussian process regression (GPR). The association between BrainAGE prediction in ASD and clinical performance was investigated. We further explored the related inter-regional profiles of gene expression from the Allen Human Brain Atlas with variability differences in the brain morphology between groups.

RESULTS

The PBSI score of GMV was negatively related to age regardless of the sample group, and the PBSI score was significantly lower in ASDs than in NCs. The regional contributions to the PBSI score of 126 brain regions in ASDs showed significant differences compared to NCs. RVR model achieved the best performance for predicting brain age. Higher inter-individual brain morphology variability was related to increased brain age, specific to communication symptoms. A total of 430 genes belonging to various pathways were identified as associated with brain cortical morphometric variation. The pathways, including short-term memory, regulation of system process, and regulation of nervous system process, were dominated mainly by gene sets for manno midbrain neurotypes.

LIMITATIONS

There is a sample mismatch between the gene expression data and brain imaging data from ABIDE. A larger sample size can contribute to the model training of BrainAGE and the validation of the results.

CONCLUSIONS

ASD has personalized heterogeneity brain morphology. The brain age gap estimation and transcription-neuroimaging associations derived from this trait are replenished in an additional direction to boost the understanding of the ASD brain.

Modern Foreign Approaches to the Assessment of Different Linguistic Levels in Children with Autism Spectrum Disorders

<p>The article presents an overview of studies on the problem of speech evaluation in childhood. We analyzed the material based on the model that describes 4 levels of language: phonetics and phonology, morphosyntax, semantics and pragmatics. The article contains the analysis of studies describing each level of language as well as existing approaches to its assessment, and reviews studies on language impairments in children with autism spectrum disorders. Additionally, the article contains a detailed analysis of existing foreign and Russian standardized methods for assessing speech development, showing which level of language organization (among each of the mentioned above methods) allows us to assess. It also describes the assessment procedures and evaluation. The review systematizes the available data on the approaches to atypical speech development in children with autism spectrum disorders.</p>