1. Aishworiya R, Ma VK, Stewart S, Hagerman R, Feldman HM. Meta-analysis of the Modified Checklist for Autism in Toddlers, Revised/Follow-up for Screening. Pediatrics. 2023.

CONTEXT: The Modified Checklist for Autism in Toddlers, Revised with Follow-up (M-CHAT-R/F) is used worldwide to screen for autism spectrum disorder (ASD). OBJECTIVE: To calculate psychometric properties of the M-CHAT-R/F for subsequent diagnosis of ASD. DATA SOURCES: Systematic searches of Medline, Embase, SCOPUS, and Trip Pro databases from January 2014 to November 2021. STUDY SELECTION: Studies were included if they (1) used the M-CHAT-R/F (2) applied standard scoring protocol, (3) used a diagnostic assessment for ASD, and (4) reported at least 1 psychometric property of the M-CHAT-R/F. DATA EXTRACTION: Two independent reviewers completed screening, full-text review, data extraction, and quality assessment, following Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines. A random-effects model was used to derive pooled estimates and assess for between-study heterogeneity. RESULTS: Of 667 studies identified, 15 with 18 distinct samples from 10 countries (49 841 children) were used in the meta-analysis. Pooled positive predictive value (PPV), was 57.7% (95% confidence interval [CI] 48.6-66.8, τ2 = 0.031). PPV was higher among high-risk (75.6% [95% CI 66.0-85.2]) than low-risk samples (51.2% [95% CI 43.0-59.5]). Pooled negative predictive value was 72.5% (95% CI 62.5-82.4 τ2 = 0.031), sensitivity was 82.6% (95% CI 76.2-88.9) and specificity 45.7% (95% CI 25.0-66.4). LIMITATIONS: Negative predictive value, sensitivity, and specificity were calculated based on small sample sizes because of limited or no evaluation of screen-negative children. CONCLUSIONS: These results support use of the M-CHAT-R/F as a screening tool for ASD. Caregiver counseling regarding likelihood of an ASD diagnosis after positive screen should acknowledge the moderate PPV.

Lien vers le texte intégral (Open Access ou abonnement)

2. Alves CL, Toutain T, de Carvalho Aguiar P, Pineda AM, Roster K, Thielemann C, Porto JAM, Rodrigues FA. Diagnosis of autism spectrum disorder based on functional brain networks and machine learning. Scientific reports. 2023; 13(1): 8072.

Autism is a multifaceted neurodevelopmental condition whose accurate diagnosis may be challenging because the associated symptoms and severity vary considerably. The wrong diagnosis can affect families and the educational system, raising the risk of depression, eating disorders, and self-harm. Recently, many works have proposed new methods for the diagnosis of autism based on machine learning and brain data. However, these works focus on only one pairwise statistical metric, ignoring the brain network organization. In this paper, we propose a method for the automatic diagnosis of autism based on functional brain imaging data recorded from 500 subjects, where 242 present autism spectrum disorder considering the regions of interest throughout Bootstrap Analysis of Stable Cluster map. Our method can distinguish the control group from autism spectrum disorder patients with high accuracy. Indeed the best performance provides an AUC near 1.0, which is higher than that found in the literature. We verify that the left ventral posterior cingulate cortex region is less connected to an area in the cerebellum of patients with this neurodevelopment disorder, which agrees with previous studies. The functional brain networks of autism spectrum disorder patients show more segregation, less distribution of information across the network, and less connectivity compared to the control cases. Our workflow provides medical interpretability and can be used on other fMRI and EEG data, including small data sets.

Lien vers le texte intégral (Open Access ou abonnement)

3. Ben Said M, Robel L, Kpenou F, Jais JP, Speranza M. Observational Cohort Study Dedicated to Autism Spectrum Disorder: Milestone Steps, Results Updates, Perspectives. Studies in health technology and informatics. 2023; 302: 716-20.

Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by persistent difficulties in two domains: social communication and interaction, alongside with restricted, repetitive pattern of behaviors. It affects children and persists into adolescence and adulthood. Its causes and underlying psychopathological mechanisms are unknown and remain to be discovered. TEDIS cohort study developed over the decade 2010-2022, in Ile-de-France region, includes 1300 patients’ files up to date, with valuable health information drawn from ASD evaluation. It provides researchers and decision makers with reliable data source to improve knowledge and practice in the context of ASD patients.

Lien vers le texte intégral (Open Access ou abonnement)

4. Colombi C, Chericoni N, Bargagna S, Costanzo V, Devescovi R, Lecciso F, Pierotti C, Prosperi M, Contaldo A. Case report: Preemptive intervention for an infant with early signs of autism spectrum disorder during the first year of life. Frontiers in psychiatry. 2023; 14: 1105253.

Autism spectrum disorder (ASD) includes neurodevelopmental conditions traditionally considered to bring life long disabilities, severely impacting individuals and their families. Very early identification and intervention during the very first phases of life have shown to significantly diminish symptom severity and disability, and improve developmental trajectories. Here we report the case of a young child showing early behavioral signs of ASD during the first months of life, including diminished eye contact, reduced social reciprocity, repetitive movements. The child received a pre-emptive parent mediated intervention based on the Infant Start, an adaptation of the Early Start Denver Model (ESDM), specifically developed for children with ASD signs during the first year of life. The child here described received intervention from 6 to 32 months of age, in combination with educational services. Diagnostic evaluations performed at several time points (8, 14, 19, and 32 months) showed progressive improvements in his developmental level and ASD symptoms. Our case study supports the possibility of identifying ASD symptoms and providing services as soon as concerns emerge even during the first year of life. Our report, in combination with recent infant identification and intervention studies, suggests the need for very early screening and preemptive intervention to promote optimal outcomes.

Lien vers le texte intégral (Open Access ou abonnement)

5. Dorsey SG, Mocci E, Lane MV, Krueger BK. Rapid effects of valproic acid on the fetal brain transcriptome: Implications for brain development and autism. bioRxiv : the preprint server for biology. 2023.

There is an increased incidence of autism among the children of women who take the anti-epileptic, mood stabilizing drug, valproic acid (VPA) during pregnancy; moreover, exposure to VPA in utero causes autistic-like symptoms in rodents and non-human primates. Analysis of RNAseq data obtained from fetal mouse brains 3 hr after VPA administration revealed that VPA significantly [p(FDR) ≤ 0.025] increased or decreased the expression of approximately 7,300 genes. No significant sex differences in VPA-induced gene expression were observed. Expression of genes associated with neurodevelopmental disorders such as autism as well as neurogenesis, axon growth and synaptogenesis, GABAergic, glutaminergic and dopaminergic synaptic transmission, perineuronal nets, and circadian rhythms was dysregulated by VPA. Moreover, expression of 400 autism risk genes was significantly altered by VPA as was expression of 247 genes that have been reported to play fundamental roles in the development of the nervous system, but are not linked to autism by GWAS. The goal of this study was to identify mouse genes that are: (a) significantly up- or down-regulated by VPA in the fetal brain and (b) known to be associated with autism and/or to play a role in embryonic neurodevelopmental processes, perturbation of which has the potential to alter brain connectivity in the postnatal and adult brain. The set of genes meeting these criteria provides potential targets for future hypothesis-driven approaches to elucidating the proximal underlying causes of defective brain connectivity in neurodevelopmental disorders such as autism.

Lien vers le texte intégral (Open Access ou abonnement)

6. Furukawa S, Nomura J, Hanafusa H, Maegawa H, Takumi T. Germ-cell-specific transcriptome analysis illuminates the chromatin and ubiquitin pathway in autism spectrum disorders. Autism research : official journal of the International Society for Autism Research. 2023.

Accumulating epidemiological studies have suggested a positive association between advanced paternal age at conception and the increased risk of neurodevelopmental outcomes such as autism spectrum disorder (ASD) in their children. Recent biological studies using human sperm have identified increased de novo mutations in aged fathers, and hyper- or hypomethylation has been identified in the sperm from aged rodents. Dysregulation of DNA methylation in sperm may explain the transgenerational effects on the pathogenesis of ASD. However, compared to these epigenetic changes in the sperm of aged males, the effects of inherited predisposition from germ cells are largely unknown. Here, we use single-cell transcriptome data sets from 13 cell lines, including 12 ASD-associated CNVs models and control, that are performed neural differentiation from mouse embryonic stem cells. This study performed comprehensive bioinformatic analyses such as gene ontology (GO), network, pathway, and upstream regulator analyses. Through these analyses, we identify several susceptible pathways, such as chromatin and ubiquitin, in addition to translational and oxidative phosphorylation. Our results suggest that dysregulation of epigenetic chromosome remodeling and ubiquitin-proteasome pathway in the germ cell is a possible modulator for subsequent differentiated cells, sperm, and egg, as a risk factor for the neurodevelopmental disorder.

Lien vers le texte intégral (Open Access ou abonnement)

7. Gabarron E, Dorronzoro E, Reichenpfader D, Denecke K. What Do Autistic People Discuss on Twitter? An Approach Using BERTopic Modelling. Studies in health technology and informatics. 2023; 302: 403-7.

Social media provide easy ways to autistic individuals to communicate and to make their voices heard. The objective of this paper is to identify the main themes that are being discussed by autistic people on Twitter. We collected a sample of tweets containing the hashtag #ActuallyAutistic during the period 10/02/2022 and 14/09/2022. To identify the most discussed topics, BERTopic modelling was applied. We manually grouped the detected topics into 6 major themes using inductive content analysis: 1) General aspects of autism and experiences of autistic individuals; 2) Autism awareness, pride and funding; 3) Interventions, mostly related to Applied Behavior Analysis; 4) Reactions and expressions; 5) Everyday life as an autistic (lifelong condition, work, housing…); and 6) Symbols and characteristics. The majority of tweets were presenting general aspects and experiences as autistic individuals; raising awareness; and about their dissatisfaction with some interventions. The identification of autistic individuals’ main discussion themes could help to develop meaningful public health agendas and research involving and addressed to autistic individuals.

Lien vers le texte intégral (Open Access ou abonnement)

8. Govarthan PK, Sinha K, Mukherjee S, Agastinose Ronickom JF. Differential Gene Expression Data Analysis of ASD Using Random Forest. Studies in health technology and informatics. 2023; 302: 1047-51.

Autism spectrum disorder (ASD) is a developmental disability caused by differences in the brain regions. Analysis of differential expression (DE) of transcriptomic data allows for genome-wide analysis of gene expression changes related to ASD. De-novo mutations may play a vital role in ASD, but the list of genes involved is still far from complete. Differentially expressed genes (DEGs) are treated as candidate biomarkers and a small set of DEGs might be identified as biomarkers using either biological knowledge or data-driven approaches like machine learning and statistical analysis. In this study, we employed a machine learning-based approach to identify the differential gene expression between ASD and Typical Development (TD). The gene expression data of 15 ASD and 15 TD were obtained from the NCBI GEO database. Initially, we extracted the data and used a standard pipeline to pre-process the data. Further, Random Forest (RF) was used to discriminate genes between ASD and TD. We identified the top 10 prominent differential genes and compared them with the statistical test results. Our results show that the proposed RF model yields 5-fold cross-validation accuracy, sensitivity and specificity of 96.67%. Further, we obtained precision and F-measure scores of 97.5% and 96.57%, respectively. Moreover, we found 34 unique DEG chromosomal locations having influential contributions in identifying ASD from TD. We have also identified chr3:113322718-113322659 as the most significant contributing chromosomal location in discriminating ASD and TD. Our machine learning-based method of refining DE analysis is promising for finding biomarkers from gene expression profiles and prioritizing DEGs. Moreover, our study reported top 10 gene signatures for ASD may facilitate the development of reliable diagnosis and prognosis biomarkers for screening ASD.

Lien vers le texte intégral (Open Access ou abonnement)

9. Harris HA, Derks IPM, Prinzie P, Louwerse A, Hillegers MHJ, Jansen PW. Interrelated development of autism spectrum disorder symptoms and eating problems in childhood: a population-based cohort. Frontiers in pediatrics. 2023; 11: 1062012.

Eating problems, such as food selectivity or picky eating, are thought to be an epiphenomenon of autism spectrum disorders (ASD). Yet eating problems are also common in the general pediatric population and overlap with ASD symptoms. However, the temporal association between ASD symptoms and eating problems is poorly understood. This study examines the bidirectional association between ASD symptoms and eating problems across child development, and investigates whether these associations differ by child sex. Participants (N = 4,930) were from the population-based Generation R Study. Parents reported their child’s ASD symptoms and eating problems using the Child Behavior Checklist at 5 assessments from toddlerhood to adolescence (1.5 to 14 years, 50% girls). A Random Intercept Cross-Lagged Panel Model was used to examine the lagged associations between ASD symptoms and eating problems at the within-person level, controlling for stable, trait-like differences at the between-person level. At the between-person level, there was a strong correlation between ASD symptoms and eating problems (β = .48, 95% CI: 0.38 to 0.57). Controlling for these between-person effects, there was limited evidence for consistent, predictive effects of ASD symptoms and eating problems at the within-person level. Associations did not differ by child sex. Findings suggest that ASD symptoms and eating problems may represent a cluster of traits that are highly stable from early childhood to adolescence, which have a minimal reciprocal effect at the individual-level. Future research could focus on these trait-like qualities to inform the development of supportive, family-focused interventions.

Lien vers le texte intégral (Open Access ou abonnement)

10. Ma K, Taylor C, Williamson M, Newton SS, Qin L. Diminished activity-dependent BDNF signaling differentially causes autism-like behavioral deficits in male and female mice. Frontiers in psychiatry. 2023; 14: 1182472.

Autism spectrum disorder (ASD) is a group of neurodevelopmental disorders with strong genetic heterogeneity and more prevalent in males than females. Recent human genetic studies have identified multiple high-risk genes for ASD, which produce similar phenotypes, indicating that diverse genetic factors converge to common molecular pathways. We and others have hypothesized that activity-dependent neural signaling is a convergent molecular pathway dysregulated in ASD. However, the causal link between diminished activity-dependent neural signaling and ASD remains unclear. Brain-derived neurotrophic factor (BDNF) is a key molecule mediating activity-dependent neural signaling. We therefore hypothesize that diminished activity-dependent BDNF signaling could confer autism-like behavioral deficits. Here, we investigated the effect of diminished activity-dependent BDNF signaling on autism-like behavioral deficits by using mice with genetic knock-in of a human BDNF methionine (Met) allele, which has decreased activity-dependent BDNF release without altering basal BDNF level. Compared with wild-type (WT) controls, diminished activity-dependent BDNF signaling similarly induced anxiety-like behaviors in male and female mice. Notably, diminished activity-dependent BDNF signaling differentially resulted in autism-like social deficits and increased self-grooming in male and female mice, and male mice were more severe than female mice. Again, sexually dimorphic spatial memory deficits were observed in female BDNF(+/Met) mice, but not in male BDNF(+/Met) mice. Our study not only reveals a causal link between diminished activity-dependent BDNF signaling and ASD-like behavioral deficits, but also identifies previously underappreciated sex-specific effect of diminished activity-dependent BDNF signaling in ASD. These mice with genetic knock-in of the human BDNF Met variant provide a distinct mouse model for studying the cellular and molecular mechanisms underlying diminished activity-dependent neural signaling, the common molecular pathway dysregulated in ASD.

Lien vers le texte intégral (Open Access ou abonnement)

11. Mantzalas J, Richdale AL, Dissanayake C. Examining subjective understandings of autistic burnout using Q methodology: A study protocol. PloS one. 2023; 18(5): e0285578.

BACKGROUND: Early research indicates that autistic burnout is a chronic, debilitating condition experienced by many autistic people across the lifespan that can have severe consequences for their mental health, wellbeing, and quality of life. To date, studies have focused on the lived experiences of autistic adults, and findings suggest that a lack of support, understanding, and acceptance by others can contribute to the risk of autistic burnout. The study outlined in this protocol will investigate how autistic people with and without experience of autistic burnout, their families, friends, healthcare professionals and non-autistic people understand the construct of autistic burnout to identify commonalities and gaps in knowledge. STUDY AND DESIGN: Q methodology will be used to investigate participants’ subjective understandings of autistic burnout. Q methodology is a mixed-methods design that is well-suited to exploratory research and can elucidate a holistic and comprehensive representation of multiple perspectives about a topic. Participants will complete a card sorting activity to rank how strongly they agree or disagree with a set of statements about autistic burnout and participate in a semi-structured interview to discuss their responses. A first-order factor analysis will be conducted for each participant group, followed by second-order factor analysis to compare the groups’ viewpoints. The interview data will provide additional insights into the factors. CONCLUSION: Q methodology has not previously been used to examine autistic and non-autistic people’s perspectives about autistic burnout. Projected study outcomes include enhanced understanding of the characteristics, risks, and protective factors of autistic burnout. The findings will have practical implications for improving detection of autistic burnout and identifying strategies to support autistic adults with prevention and recovery. The results may also inform the development of a screening protocol and identify potential avenues for future research.

Lien vers le texte intégral (Open Access ou abonnement)

12. Ni Z, Qian Y, Li H, Yao Z. Predicting Family Implementation of Complementary and Alternative Medicine in Autism Online Communities. Studies in health technology and informatics. 2023; 302: 123-4.

Complementary and alternative medicine (CAM) is widely adopted by families with autistic children. This study aims to predict family caregivers’ CAM implementation in Autism online communities. Dietary interventions were reported as a case study. We extracted behavioral (degree and betweenness), environmental (positive feedback and social persuasion), and personal features (language style) of family caregivers in online communities. The results of the experiment showed that random forests performed well in predicting families’ tendency to implement CAM (AUC=0.887). It is promising to use machine learning to predict and intervene in the CAM implementation by family caregivers.

Lien vers le texte intégral (Open Access ou abonnement)

13. Ranjan R, Jain M, Kumar P, Sethi G, Singh J. Exploring the path to pathos « Lived experiences of parents of children with autism spectrum disorder »: An interpretative phenomenological analysis. Indian journal of psychiatry. 2023; 65(3): 310-8.

BACKGROUND: Children with autism spectrum disorder (ASD) require lifetime support by the family, thus posing a great amount of stress among parents. Understanding lived experiences of parents who provide lifelong support will guide in planning effective treatment for children with ASD. In view of this, the study was aimed to depict and understand the lived experiences of parents of children with ASD and making sense of it. METHODS: This interpretative phenomenological analysis research design was carried out on 15 parents of children with ASD coming to the tertiary care referral hospital of eastern zone of India. In-depth interviews were conducted to understand the lived experiences of parents. RESULTS: The current study identified six themes: major symptom recognition; myths, beliefs, and stigma related to children with ASD; help seeking behavior; coping with challenging experiences; support system; uncertainties, insecurities, and gleam of hope. CONCLUSION: Lived experiences were found to be predominantly difficult for most of the parents of children with ASD, and inadequate services pose a major challenge to them. The findings highlight the need for involving the parents in the treatment programs as early as possible or extending appropriate support to the family.

Lien vers le texte intégral (Open Access ou abonnement)

14. Schmitt LM, Will M, Shaffer R, Erickson C. A Paradigm Shifting View of Intellectual Disability: A Near Normal Distribution of IQ in Fragile X Syndrome. Research square. 2023.

Fragile X Syndrome (FXS) is an X-linked disorder leading to the loss of expression of FMR1 -protein product, FMRP. The absence or deficiency of FMRP is thought to result in the characteristic FXS phenotypes, including intellectual disability. Identifying the relationship between FMRP levels and IQ may be critical to better understand underlying mechanisms and advance treatment development and planning. A sample of 80 individuals with FXS (67% male), aged 8-45 years, completed IQ testing and blood draw via venipuncture to determine the relationship between IQ scores and FMRP levels as well as the normalcy of IQ distributions. In females with FXS only, higher FMRP levels were associated with higher IQ. In contrast, males with FXS showed a downward shifted but otherwise normal distribution of IQ scores. Our findings offer a paradigm-shifting views of FXS-males with FXS have normally distributed IQ that is downshifted 5 standard deviations. Our novel work provides evidence of a « FXS standard curve », and is a critical step towards establishing molecular markers of disease severity in FXS. There is much future work to better understand the mechanism by which FMRP loss leads to intellectual disability and what biological/genetic and socio-environmental factors contribute to variation in IQ.

Lien vers le texte intégral (Open Access ou abonnement)

15. Soleiman P, Moradi H, Mehralizadeh B, Ameri H, Arriaga RI, Pouretemad HR, Baghbanzadeh N, Vahid LK. Fully robotic social environment for teaching and practicing affective interaction: Case of teaching emotion recognition skills to children with autism spectrum disorder, a pilot study. Frontiers in robotics and AI. 2023; 10: 1088582.

21st century brought along a considerable decrease in social interactions, due to the newly emerged lifestyle around the world, which became more noticeable recently of the COVID-19 pandemic. On the other hand, children with autism spectrum disorder have further complications regarding their social interactions with other humans. In this paper, a fully Robotic Social Environment (RSE), designed to simulate the needed social environment for children, especially those with autism is described. An RSE can be used to simulate many social situations, such as affective interpersonal interactions, in which observational learning can take place. In order to investigate the effectiveness of the proposed RSE, it has been tested on a group of children with autism, who had difficulties in emotion recognition, which in turn, can influence social interaction. An A-B-A single case study was designed to show how RSE can help children with autism recognize four basic facial expressions, i.e., happiness, sadness, anger, and fear, through observing the social interactions of two robots speaking about these facial expressions. The results showed that the emotion recognition skills of the participating children were improved. Furthermore, the results showed that the children could maintain and generalize their emotion recognition skills after the intervention period. In conclusion, the study shows that the proposed RSE, along with other rehabilitation methods, can be effective in improving the emotion recognition skills of children with autism and preparing them to enter human social environments.

Lien vers le texte intégral (Open Access ou abonnement)

16. Waddington H, Minnell H, Patrick L, van Der Meer L, Monk R, Woods L, Whitehouse AJ. Community perspectives on the appropriateness and importance of support goals for young autistic children. Autism : the international journal of research and practice. 2023: 13623613231168920.

Researchers do not know much about what autistic adults, parents and professionals think about support goals for young autistic children. People’s views of support goals might also be influenced by their beliefs about early support more generally. This survey involved 87 autistic adults, 159 parents of autistic children and 80 clinical professionals living in New Zealand and Australia. We asked participants questions about themselves and what they thought about early support for young autistic children in general. We then asked participants to rate whether different support goals were appropriate for young autistic children and, if they were appropriate, to rate their level of priority. We found that autistic adults, parents and professionals all rated goals about the adult changing to better support the child, reducing and replacing harmful behaviours and improving the child’s quality of life as the highest priorities. They all rated goals about autism characteristics, play skills and academic skills as the lowest priorities. Compared to parents and/or professionals, autistic adults gave lower priority ratings for play skills, autism characteristics and participation goals. Autistic adults were also more likely to rate goals related to play skills and autism characteristics as inappropriate. While these three participant groups generally agreed on the order of priority of early support goals for young autistic children, autistic adults found goals related to autism characteristics, play and/or participation to be an even lower priority and less appropriate than parents and professionals.

Lien vers le texte intégral (Open Access ou abonnement)

17. Yang B, Wang M, Zhou W, Wang X, Chen S, Yuan LX, Dong GH. Edge-centric functional network analyses reveal disrupted network configuration in autism spectrum disorder. Journal of affective disorders. 2023.

BACKGROUND: Neuroscientific evidence suggests that the pathological symptoms associated with autism spectrum disorders (ASD) are not confined to a single brain region but involve networks of the brain on a larger spatial scale. Analyzing diagrams of edge-edge interactions could provide important perspectives on the organization and function of complex systems. METHODS: Resting-state fMRI data from 238 ASD patients and 311 healthy controls (HCs) were included in the current study. We used the thalamus as the mediating node to calculate the edge functional connectivity (eFC) of the brain network and compared the ASD subjects and HCs. RESULTS: Compared with the HCs, the ASD subjects exhibited abnormalities in the central node thalamus and four brain regions (amygdala, nucleus accumbens, pallidum and hippocampus), as well as in the eFC formed by the inferior frontal gyrus (IFG) (or middle temporal gyrus (MTG)). In addition, ASD subjects showed variable characteristics of the eFC between nodes in different networks. CONCLUSIONS: The changes in these brain regions may be due to the disturbance in the reward system, which leads to coherence in the instantaneous comovement of the functional connections formed by these brain regions in ASD. This notion also reveals a functional network feature between the cortical and subcortical regions in ASD.

Lien vers le texte intégral (Open Access ou abonnement)

18. Zhu H, Wang J, Zhao YP, Lu M, Shi J. Contrastive Multi-View Composite Graph Convolutional Networks Based on Contribution Learning for Autism Spectrum Disorder Classification. IEEE transactions on bio-medical engineering. 2023; 70(6): 1943-54.

The resting-state functional magnetic resonance imaging (rs-fMRI) faithfully reflects the brain activities and thus provides a promising tool for autism spectrum disorder (ASD) classification. Up to now, graph convolutional networks (GCNs) have been successfully applied in rs-fMRI based ASD classification. However, most of these methods were developed based on functional connectivities (FCs) that only reflect low-level correlation between brain regions, without integrating both high-level discriminative knowledge and phenotypic information into classification. Besides, they suffered from the overfitting problem caused by insufficient training samples. To this end, we propose a novel contrastive multi-view composite GCN (CMV-CGCN) for ASD classification using both FCs and HOFCs. Specifically, a pair of graphs are constructed based on the FC and HOFC features of the subjects, respectively, and they share the phenotypic information in the graph edges. A novel contrastive multi-view learning method is proposed based on the consistent representation of both views. A contribution learning mechanism is further incorporated, encouraging the FC and HOFC features of different subjects to have various contribution in the contrastive multi-view learning. The proposed CMV-CGCN is evaluated on 613 subjects (including 286 ASD patients and 327 NCs) from the Autism Brain Imaging Data Exchange (ABIDE). We demonstrate the performance of the method for ASD classification, which yields an accuracy of 75.20% and an area under the curve (AUC) of 0.7338. Experimental results show that our proposed method outperforms state-of-the-art methods on the ABIDE database.

Lien vers le texte intégral (Open Access ou abonnement)