1. Alnafjan A, Alghamdi M, Alhakbani N, Al-Ohali Y. Improving Imitation Skills in Children with Autism Spectrum Disorder Using the NAO Robot and a Human Action Recognition. Diagnostics (Basel);2024 (Dec 29);15(1)

Background/Objectives: Autism spectrum disorder (ASD) is a group of developmental disorders characterized by poor social skills, low motivation in activities, and a lack of interaction with others. Traditional intervention approaches typically require support under the direct supervision of well-trained professionals. However, teaching and training programs for children with ASD can also be enhanced by assistive technologies, artificial intelligence, and robotics. Methods: In this study, we examined whether robotics can improve the imitation skills of children with autism and support therapists during therapeutic sessions. We designed scenarios for training hand clapping imitation skills using the NAO robot and analyzed the interaction between children with autism and the robot. Results: We developed a deep learning approach based on the human action recognition algorithm for analyzing clapping imitation. Conclusions: Our findings suggest that integrating robotics into therapeutic practices can effectively enhance the imitation skills of children with ASD, offering valuable support to therapists.

Lien vers le texte intégral (Open Access ou abonnement)

2. Bent C, Dwyer P, Hudry K. Optimal amount of support for young Autistic children depends on their individual and family needs. Evid Based Nurs;2025 (Jan 11)

Lien vers le texte intégral (Open Access ou abonnement)

3. Chi IJ, Tsai SJ, Chen CH, Yang AC. Identifying Distinct Developmental Patterns of Brain Complexity in Autism: A Cross-Sectional Cohort Analysis Using the Autism Brain Imaging Data Exchange. Psychiatry Clin Neurosci;2025 (Jan 11)

AIM: Autistic traits exhibit neurodiversity with varying behaviors across developmental stages. Brain complexity theory, illustrating the dynamics of neural activity, may elucidate the evolution of autistic traits over time. Our study explored the patterns of brain complexity in autistic individuals from childhood to adulthood. METHODS: We analyzed functional magnetic resonance imaging data from 1087 autistic participants and neurotypical controls aged 6 to 30 years within the ABIDE I (Autism Brain Imaging Data Exchange) data set. Sample entropy was calculated to measure brain complexity among 90 brain regions, utilizing an automated anatomical labeling template for voxel parcellation. Participants were grouped using sliding age windows with partial overlaps. We assessed the average brain complexity of the entire brain and brain regions for both groups across age categories. Cluster analysis was conducted using generalized association plots to identify brain regions with similar developmental complexity trajectories. Finally, the relationship between brain region complexity and autistic traits was examined. RESULTS: Autistic individuals may tend toward higher whole-brain complexity during adolescence and lower complexity during childhood and adulthood, indicating possible distinct developmental trajectories. However, these results do not remain after Bonferroni correction. Two clusters of brain regions were identified, each with unique patterns of complexity changes over time. Correlations between brain region complexity, age, and autistic traits were also identified. CONCLUSION: The study revealed brain complexity trajectories in autistic individuals, providing insight into the neurodiversity of autism and suggesting that age-related changes in brain complexity could be a potential neurodevelopmental marker for the dynamic nature of autism.

Lien vers le texte intégral (Open Access ou abonnement)

4. de Oliveira LC, Janizello GP, Santos IS, de Oliveira N, Carvalho DB, Napoli SB, Schiariti V, Gomes C, Silva SM. Systematic review of outcome measures used in support programs designed to enhance the functioning for autistic children and adolescents and ICF content mapping. Disabil Rehabil;2025 (Jan 11):1-25.

PURPOSE: 1) To identify outcome measures used in support programs designed to enhance functioning in autistic children and adolescents, and 2) To map the content of these measures to the domains of the International Classification of Functioning, Disability and Health (ICF). METHODS: A systematic review was conducted. Searches were performed in Medline/PubMed, EMBASE and Virtual Health Library databases, with no restrictions imposed regarding language or year of publication. Studies that used outcome measures to assess functioning and/or disability in autistic individuals up to 18 years of age were included. RESULTS: A total of 20 outcome measures were identified. The Vineland Adaptive Behavior Scales was the most used outcome measure. The most frequently associated ICF domains were d7 « Interpersonal interactions and relationships, » d3 « Communication, » and b1 « Mental functions. » The most extensively assessed component was activities and participation. Bladder and bowel control (b5 « digestive, metabolic, and endocrine functions » and b6 « genitourinary and reproductive functions »), multitasking (d2 « general tasks and demands »), and environmental factors (e2 « natural environment and human-made changes » and e4 « attitudes ») were each linked only once to the outcome measures. None of the studies addressed the body structures component. CONCLUSION: No single instrument adequately encompasses all ICF domains, underscoring a significant gap in current assessment tools. The domains of body structures, environmental factors and some body functions were under-assessed in studies involving autistic children and adolescents. No single instrument adequately encompasses all ICF domains, underscoring a significant gap in current assessment tools.The domains of body structures, environmental factors and some body functions were under-assessed in studies involving autistic children and adolescents.Future interventions must incorporate a robust evaluation of environmental factors and body functions, as these components have been frequently under-assessed in intervention studies.There is an urgent need to develop and validate comprehensive assessment tools that integrate all ICF domains to ensure a thorough approach in the rehabilitation of autistic children and adolescents. eng

Lien vers le texte intégral (Open Access ou abonnement)

5. Dellazari L, de Bem É B, Falcão AB, Manjabosco FM, Sorato GB, Berto LF, Dantas VA, Da Rosa A, Graeff-Martins AS, Kieling RR, Salum Junior GA, Rohde LA, Caye A. Mind the gap: the inconceivable void on the epidemiology of autism spectrum disorders in Brazil. Braz J Psychiatry;2025 (Jan 11)

Lien vers le texte intégral (Open Access ou abonnement)

6. Dong Y, Mo J, Gong B, Jin R, Zheng H, Chow BW. Effects of Using Prompts During Parent-Child Shared Reading on the Language Development of Mildly Autistic Children. J Autism Dev Disord;2025 (Jan 11)

The use of literal prompts (LPs) and inferential prompts (IPs) in shared book reading (SBR) facilitates children’s use of language and promotes their thinking and understanding about the stories discussed and beyond. Furthermore, SBR provides a platform for mildly autistic children to have multiple rounds of communication with educators. This study investigated the contribution of LPs and IPs on the language development and affective factors of language learning in mildly autistic children. This study included 187 mildly autistic Chinese children who were stratified by random sampling and assigned into three groups (LP, IP and control). The mildly autistic children’s language skills were tested immediately before and after the 12-week SBR intervention. Their parents were also included in this study. The results indicated that using prompts had positive effects on the mildly autistic children’s language skills and on the affective factors central to language development. Moreover, LPs were beneficial in fostering mildly autistic children’s affective factor development, whilst IPs fostered their Chinese word reading and listening comprehension skills. These findings indicated the benefits of using prompts during parent-child SBR, along with the extent to which prompts contribute to different language skills and affective factors central to language development in mildly autistic children.

Lien vers le texte intégral (Open Access ou abonnement)

7. Goscicki BL, Scoggins ME, Espinosa GH, Hodapp RM. A « Round, Bruising Sort of Pain »: Autistic Girls’ Social Camouflaging in Inclusive High School Settings. J Autism Dev Disord;2025 (Jan 10)

Although autistic females often « camouflage » their autism, few studies examine the degree to which adolescent females demonstrate these behaviors in inclusive school settings. We examined: (a) the nature, extent, and underlying motivation of camouflaging in high school; (b) the extent to which autistic girls’ characteristics related to camouflaging settings, people, benefits, costs, and school supports; and (c) how girls’ open-ended descriptions agreed with closed-ended camouflaging ratings. Using quantitative and qualitative analyses, this study examined the extent, domains, costs, and benefits of autistic females’ school-based camouflaging. Thirty-one autistic female adolescents, all included in general education classrooms, answered rating and interview questions. Autistic females camouflaged most often in general education classrooms and with teachers and neurotypical peers that they did not know well; least often at home or with neurodivergent friends. Later age of diagnosis was associated with more camouflaging and camouflaging costs. Qualitative analyses revealed four themes: autistic identity; negative peer experiences; negative consequences of camouflaging; and value of neurodivergent friends. Some qualitative findings converged with quantitative findings, others diverged. Implications are discussed for research and practice for supporting autistic females in general education school settings.

Lien vers le texte intégral (Open Access ou abonnement)

8. Hansen N, Dischler A, Dias C. Beyond the Synapse: FMR1 and FMRP Molecular Mechanisms in the Nucleus. Int J Mol Sci;2024 (Dec 30);26(1)

FMR1 (Fragile X messenger ribonucleoprotein 1), located on the X-chromosome, encodes the multi-functional FMR1 protein (FMRP), critical to brain development and function. Trinucleotide CGG repeat expansions at this locus cause a range of neurological disorders, collectively referred to as Fragile X-related conditions. The most well-known of these is Fragile X syndrome, a neurodevelopmental disorder associated with syndromic facial features, autism, intellectual disabilities, and seizures. However, CGG expansions of different sizes also confer a risk of neuropsychiatric and neurodegenerative disorders throughout the lifespan, through distinct molecular mechanisms. Although Fragile X syndrome is associated with downstream synaptic deficits and neuronal hyperexcitability, work in the past decade has demonstrated that both the causative FMR1 trinucleotide repeat expansion and FMRP itself play important roles in nuclear function and regulation, including non-canonical nucleic acid structure formation and chromatin dynamics. These effects are critical to cellular pathophysiology, although the full extent of their contribution to clinical phenotypes is only just emerging. Here, we present a focused review on some of the nuclear consequences of FMR1/FMRP dysregulation, including parallels in other repeat expansion disorders, ranging from studies in model systems to human cells and tissues.

Lien vers le texte intégral (Open Access ou abonnement)

9. Jaradat AS, Wedyan M, Alomari S, Barhoush MM. Using Machine Learning to Diagnose Autism Based on Eye Tracking Technology. Diagnostics (Basel);2024 (Dec 30);15(1)

Background/Objectives: One of the key challenges in autism is early diagnosis. Early diagnosis leads to early interventions that improve the condition and not worsen autism in the future. Currently, autism diagnoses are based on monitoring by a doctor or specialist after the child reaches a certain age exceeding three years after the parents observe the child’s abnormal behavior. Methods: The paper aims to find another way to diagnose autism that is effective and earlier than traditional methods of diagnosis. Therefore, we used the Eye Gaze fixes map dataset and Eye Tracking Scanpath dataset (ETSDS) to diagnose Autistic Spectrum Disorder (ASDs), while a subset of the ETSDS was used to recognize autism scores. Results: The experimental results showed that the higher accuracy rate reached 96.1% and 98.0% for the hybrid model on Eye Gaze fixes map datasets and ETSDS, respectively. A higher accuracy rate was reached (98.1%) on the ETSDS used to recognize autism scores. Furthermore, the results showed the outperformer for the proposed method results compared to previous works. Conclusions: This confirms the effectiveness of using artificial intelligence techniques in diagnosing diseases in general and diagnosing autism, in addition to the need to increase research in the field of diagnosing diseases using advanced techniques.

Lien vers le texte intégral (Open Access ou abonnement)

10. Khan AA, Dewald HD. Nitric oxide and peroxynitrite as new biomarkers for early diagnosis of autism. Brain Res;2025 (Jan 8);1850:149438.

Autism spectrum disorder, or autism, is a neurodevelopmental disorder of the developing child’s brain with a genetic causality. It can be diagnosed at about three years after birth when it begins to present itself via a range of neuropsychiatric symptoms. Nitric oxide is a crucial small molecule of life synthesized within cells of our body systems, including cells of our brain. Peroxynitrite is the product of reaction between superoxide anion and nitric oxide. It normally isomerizes into harmless nitrates or nitrites. However, when excessive superoxide anion is present, the cellular concentration of peroxynitrite can increase to a toxic level. Autism has been suggested to cause oxidative damage to brain cells. Until now, it is impossible to sample tissue from a live brain. Instead, stem cells can be derived (from an autism patient’s somatic cells) which can then be differentiated and chemically directed to grow into miniature 3-dimensional tissue masses resembling specific brain regions (e.g., the cortex) called brain organoids. This review discusses utilizing nitric oxide and peroxynitrite as biomarkers and comparing their relative concentrations in stem cells and stem cell derived brain organoids of healthy and autistic individuals to develop a bioanalytical process for early diagnosis of autism.

Lien vers le texte intégral (Open Access ou abonnement)

11. Kuenzel E, Al-Saoud S, Fang M, Duerden EG. Early childhood stress and amygdala structure in children and adolescents with neurodevelopmental disorders. Brain Struct Funct;2025 (Jan 11);230(1):29.

Children and adolescents with neurodevelopmental disorders such as autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) may be more susceptible to early life stress compared to their neurotypical peers. This increased susceptibility may be linked to regionally-specific changes in the striatum and amygdala, brain regions sensitive to stress and critical for shaping maladaptive behavioural responses. This study examined early life stress and its impact on striatal and amygdala development in 62 children and adolescents (35 males, mean age = 10.12 years, SD = 3.6) with ASD (n = 14), ADHD (n = 28), or typical development (TD, n = 20) across two cohorts. We assessed stress from various sources, including from the family environment, loss of loved ones, social stress, and illness/injury. We further examined parenting styles as potential moderators of the effects of early life stress. Volumes of the striatum and amygdala were extracted using an automatic segmentation algorithm. Significant group differences in childhood stress exposure were observed (F = 3.29, df = 8, p = 0.002), with autistic children facing more early life stressors (social stress, illness/injury) compared to those with ADHD and neurotypical peers (both, p < 0.002). In autistic children, amygdala volumes were significantly associated with early life stress related to the familial environment, experiences of significant loss, and illness/injury (all, p < 0.03). Positive parenting moderated these effects. These findings suggest that autistic children are more likely to experience early life stress and exhibit region-specific changes in the amygdala, a key brain region implicated in emotional processing and stress responses. This underscores the need for targeted interventions to support autistic children in managing early life stress to potentially mitigate its impact on brain development.

Lien vers le texte intégral (Open Access ou abonnement)

12. Milane NC, Semensato MT, Pilatti LA. Research Tools for Eating Behavior in People with Autism Spectrum Disorder (ASD). J Autism Dev Disord;2025 (Jan 11)

OBJECTIVE: To identify the tools used to assess eating behaviors in individuals with Autism Spectrum Disorder (ASD) and summarize their distribution, citation rates, journal publication, JCR scores, and psychometric properties. METHODS: A literature review was conducted to identify studies on eating behavior in individuals with ASD. The search included various descriptors and combinations of keywords in databases such as Medline/PubMed, Science Direct, Scopus, SciELO, and Web of Science. The studies were filtered to focus on articles published in the last five years. Thirty-seven relevant studies were identified and analyzed to summarize the tools used, their distribution in the literature, citation rates, and psychometric properties. RESULTS: Thirty-seven relevant studies were identified. The Journal of Autism and Developmental Disorders published the most studies (5). The Brief Autism Mealtime Behavior Inventory (BAMBI) was the most frequently used instrument, appearing in 15 studies, followed by the Behavioral Pediatrics Feeding Assessment Scale (BPFAS) in 8 studies. Both instruments demonstrated solid psychometric properties, with BAMBI showing good internal consistency (α = 0.88) and BPFAS a Cronbach’s alpha of 0.82. CONCLUSION: Most of the instruments used in studies on eating behavior in individuals with ASD demonstrate satisfactory psychometric properties. BAMBI and BPFAS stand out for their widespread use but are limited to covering only specific age ranges.

Lien vers le texte intégral (Open Access ou abonnement)

13. Navas P, Arias VB, Vicente E, Esteban L, Guillén VM, Alvarado N, Heras I, Rumoroso P, García-Domínguez L, Verdugo M. Empowering lives: How deinstitutionalization and community living improve the quality of life of individuals with intellectual and developmental disabilities. Res Dev Disabil;2025 (Jan 11);157:104909.

BACKGROUND: The number of people with intellectual and developmental disabilities (IDD) living in residential settings has not changed substantially in recent years in Spain. To change this situation the project « My House: A Life in the Community » aims to promote the transition of individuals with IDD and high support needs into community settings. AIMS: This paper deepens the understanding of the underlying mechanisms responsible for the changes observed in the process of deinstitutionalization. METHODS: A longitudinal study was designed to evaluate the quality of life of individuals with IDD (n = 90) across different environments at two distinct time points: when they were living in an institution and nine months after transitioning to a community-based setting. A comparison group (n = 72) consisting of individuals who remained institutionalized was included to carry out intergroup comparisons. T-tests were used to estimate mean differences both between and within groups. Longitudinal path models were used to investigate the processes underlying the relationships between variables. RESULTS: After transition, movers obtained significantly higher mean scores on all variables with large or very large effect sizes. However, simply moving to a different place was not the factor responsible for the observed improvements: positive changes in quality of life require the constant availability of opportunities to support decision making. IMPLICATIONS: deinstitutionalization will only lead to improvements in quality of life if it favors people’s control over their lives. Community living should therefore be understood not as an autonomous life but as a chosen one.

Lien vers le texte intégral (Open Access ou abonnement)

14. Pan N, Chen L, Wu B, Chen F, Chen J, Huang S, Guo C, Wu J, Wang Y, Chen X, Yang S, Jing J, Weng X, Lin L, Liang J, Wang X. Developing a simplified measure to predict the risk of autism spectrum disorders: Abbreviating the M-CHAT-R using a machine learning approach in China. Psychiatry Res;2025 (Jan 3);344:116353.

BACKGROUND: Early screening for autism spectrum disorder (ASD) is crucial, yet current assessment tools in Chinese primary child care are limited in efficacy. OBJECTIVE: This study aims to employ machine learning algorithms to identify key indicators from the 20-item Modified Checklist for Autism in Toddlers, revised (M-CHAT-R) combining with ASD-related sociodemographic and environmental factors, to distinguish ASD from typically developing children. METHODS: Data from our prior validation study of the Chinese M-CHAT-R (August 2016-March 2017, n = 6,049 toddlers) were reviewed. We extracted the 20-item M-CHAT-R data and integrated 17 sociodemographic and environmental risk factors associated with ASD development to strengthen M-CHAT-R’s machine learning screening. Five feature selection methods were used to extract subsets from the original set. Six machine learning algorithms were applied to identify the optimal subset distinguishing clinically diagnosed ASD toddlers from typically developing toddlers. FINDINGS: Nine features were grouped into three subsets: subset 1 contained unanimously recommended items (A1 [Follows point], A3 [Pretend play], A9 [Brings objects to show], A10 [Response to name] and A16 [Gazing following]). Subset 2 added two items (A17 [Gaining parent’s attention] and A18 [Understands what is said]), and subset 3 included two more items (A8 [Interest in other children] and child’s age). The top-performing algorithm resulted in a seven-item classifier of subset 2 with 92.5 % sensitivity, 90.1 % specificity, and 10.0 % positive predictive value. CONCLUSIONS: Machine learning classifiers effectively differentiate ASD toddlers from typically developing toddlers using a reduced M-CHAT-R item set. CLINICAL IMPLICATIONS: This highlights the clinical significance of machine learning-optimized models for ASD screening in primary health care centers and broader applications.

Lien vers le texte intégral (Open Access ou abonnement)

15. Petrolo E, Guerrera S, Logrieco MG, Casula L, Vicari S, Valeri G. The role of executive functions in preschool children with autism spectrum disorder: A short narrative review. Res Dev Disabil;2025 (Jan 9);157:104905.

Atypical executive functions (EFs) are well-documented in individuals with autism spectrum disorders (ASD) across all ages. However, most research focuses on EFs impairments in school-aged children and older, with less attention to preschool children. Understanding EF deficits in this age group is challenging and underexplored due to limited studies and measurement difficulties. The current short narrative review’s aim is to provide an update on the knowledge on EFs in preschool children with ASD and their association with ASD symptoms. Despite varied results, recent research suggests early EF difficulties, potentially linked to greater impairments in social skills and externalizing behaviors. This suggests the importance of implementing early interventions that take into account the enhancement of these areas from an early age. Further investigation in this age group could enhance our understanding of ASD.

Lien vers le texte intégral (Open Access ou abonnement)

16. Robinson J, Gendelberg D, Chung A, Jimenez-Almonte JH, Khandehroo B, Anand N. Segmental Interbody, Muscle-Preserving, Ligamentotaxis-Enabled Reduction: « SIMPLER » Technique for cMIS Correction of ASD. Int J Spine Surg;2025 (Jan 10)

BACKGROUND: Correction of adult spinal deformity (ASD) through minimally invasive techniques is a challenging endeavor and has typically been reserved for experienced surgeons. This publication aims to be the first high-resolution technique guide to demonstrate a reproducible technique for ASD correction utilizing circumferential minimally invasive surgery (cMIS) without an osteotomy. The Segmental Interbody, Muscle-Preserving, Ligamentotaxis-Enabled Reduction (SIMPLER) technique is a novel ligamentotaxis-based scoliosis surgery that represents a paradigm shift from traditional osteotomies toward patient-specific correction. METHODS: The senior author’s (N.A.) cMIS technique for ASD correction without an osteotomy is described using high-resolution photographs, computer-generated imagery (CGI), and a case example. Step-by-step intraoperative photographs document a novel muscle-preserving posterior spinal exposure, spinal robotic safety protocol for instrumentation, dedicated deformity instrumentation system, rod reduction sequence, and minimally invasive fusion technique. CGI assists to reinforce technical considerations described by intraoperative photographs. RESULTS: The SIMPLER technique is documented from incision to closure with high-resolution pictures including CGI to highlight concepts documented in photographs. Technical considerations were detailed for all aspects involved in the planning and execution of an osteotomy-free deformity correction. CONCLUSION: This represents the first in-depth technical description of ligamentotaxis-based, osteotomy-free, ASD scoliosis correction. The SIMPLER approach is reproducible and minimally invasive and can be done routinely for appropriately selected deformity candidates. This technique serves as a foundation to externally validate previously described cMIS ASD deformity correction outcomes. CLINICAL RELEVANCE: Circumferential minimally invasive spinal deformity correction is reproducible and can be achieved reliably through the use of the SIMPLER technique, without the use of an osteotomy.

Lien vers le texte intégral (Open Access ou abonnement)

17. Rojo-Marticella M, Arija V, Canals-Sans J. Effect of Probiotics on the Symptomatology of Autism Spectrum Disorder and/or Attention Deficit/Hyperactivity Disorder in Children and Adolescents: Pilot Study. Res Child Adolesc Psychopathol;2025 (Jan 11)

The aim of this study is to investigate the impact of using probiotics with strains related to dopamine and gamma-aminobutyric acid production on clinical features of autism spectrum disorder (ASD) and/or attention deficit/hyperactivity disorder (ADHD). This randomized, controlled trial involved 38 children with ADHD and 42 children with ASD, aged 5-16 years, who received probiotics (Lactiplantibacillus plantarum and Levilactobacillus brevis 109/cfu/daily) or placebo for 12 weeks. Parent-reported symptoms were assessed using Conners’ 3rd-Ed and the Social Responsiveness Scale Test, 2nd-Ed (SRS-2), and children completed the Conners Continuous Performance Test, 3rd-Ed (CPT 3) or Conners Kiddie CPT, 2nd-Ed (K-CPT 2). Executive functions, quality of life and sleep patterns were also parent-assessed. Intention-to-treat analyses, controlling for sociodemographic and nutritional covariates, revealed no significant inter-group differences in parent-reported or neuropsychological data after the probiotic intervention. However, age-stratified analyses showed improved hyperactivity-impulsivity symptoms in younger children with ASD (Cohen’s d = 1.245) and ADHD (Cohen’s d = 0.692). Intra-group analyses supported these findings in the aforementioned age and intervention group for both diagnoses. An improvement in impulsivity for children with ASD was also observed in the intra-group analysis of the CPT commissions scores (probiotic: p = 0.001, Cohen’s d = -1.216; placebo: p = 0.013, Cohen’s d = -0.721). A better comfort score (quality of life) was shown in children with ASD (probiotic: p = 0.010, Cohen’s d = 0.722; placebo: p = 0.099, Cohen’s d = 0.456). The probiotics used, may improve hyperactivity-impulsivity in children with ASD or/and ADHD and quality of life in children with ASD. Further research is warranted to explore probiotics as an adjunctive therapeutic intervention for NDs.Trial registration: clinicaltrials.gov Identifier: NCT05167110.

Lien vers le texte intégral (Open Access ou abonnement)

18. Tang J, Chen J, Hu M, Hu Y, Zhang Z, Xiao L. Diagnosis of Autism Spectrum Disorder (ASD) by Dynamic Functional Connectivity Using GNN-LSTM. Sensors (Basel);2024 (Dec 30);25(1)

Early detection of autism spectrum disorder (ASD) is particularly important given its insidious qualities and the high cost of the diagnostic process. Currently, static functional connectivity studies have achieved significant results in the field of ASD detection. However, with the deepening of clinical research, more and more evidence suggests that dynamic functional connectivity analysis can more comprehensively reveal the complex and variable characteristics of brain networks and their underlying mechanisms, thus providing more solid scientific support for computer-aided diagnosis of ASD. To overcome the lack of time-scale information in static functional connectivity analysis, in this paper, we proposes an innovative GNN-LSTM model, which combines the advantages of long short-term memory (LSTM) and graph neural networks (GNNs). The model captures the spatial features in fMRI data by GNN and aggregates the temporal information of dynamic functional connectivity using LSTM to generate a more comprehensive spatio-temporal feature representation of fMRI data. Further, a dynamic graph pooling method is proposed to extract the final node representations from the dynamic graph representations for classification tasks. To address the variable dependence of dynamic feature connectivity on time scales, the model introduces a jump connection mechanism to enhance information extraction between internal units and capture features at different time scales. The model achieves remarkable results on the ABIDE dataset, with accuracies of 80.4% on the ABIDE I and 79.63% on the ABIDE II, which strongly demonstrates the effectiveness and potential of the model for ASD detection. This study not only provides new perspectives and methods for computer-aided diagnosis of ASD but also provides useful references for research in related fields.

Lien vers le texte intégral (Open Access ou abonnement)

19. Ulaganathan S, Harshavardhanan P, Ganapathi Raju NV, Parthasarathy G. Hybrid optimization enabled DenseNet for autism spectrum disorders using MRI image. Comput Biol Chem;2024 (Dec 30);115:108335.

Autism spectrum disorder (ASD) is the neuro-developmental disorder caused by various changes in the brain. It affects the life conditions with social interaction and communication. Most of the previous researches used the various techniques for the early detection to reduce the ASD, but it had been occurred several complications such as, time expenses, and low accessibility for diagnosis.This paper aims to develop the JSTO-DenseNetmodel is for the detection of ASD. In this paper, an input autism brainimage is considered as an input applied to image pre-processing phase. In image pre-processing, the clatters are removed utilizing Gaussian filtering and also, Region of Interest (ROI) extraction is carried out. Thereafter, extraction of pivotal region is done based on functional connectivity utilizing proposed Jaya Sewing Training Optimization (JSTO). The JSTO is newly introduced by combining Jaya algorithm and Sewing Training-Based Optimization (STBO). Thus, output-1 is obtained. In feature extraction phase, grey level co-occurrence matrix (GLCM) features like entropy, correlation, energy, homogeneity, inverse difference moment, Angular second moment and texture features namelylocal ternary patterns (LTP), Local Optimal Oriented Pattern (LOOP) and Histogram of Oriented Gradients (HOG) are extracted from the Magnetic Resonance Imaging (MRI). Therefore, output-2 is obtained. From output-1 and output-2, ASD classification is accomplished using DenseNet, which is trained employing same proposed JSTO.The proposed JSTO-DenseNet model achieves the highest accuracy of 94.8 %, True Positive Rate (TPR) of 90 %, True Negative Rate (TNR) of 90.5 %, un-weighted average recall (UAR) of 89.8 % and the lowest False Negative Rate (FNR) of 86.7 %, and False Positive Rate of 82.6 %, when compared with other traditional methods like, Explainable Artificial Intelligence (XAI), Hybrid deep lightweight feature generator, CLAttention, Two stream end-to-end deep learning (DL), Auto-Encoder feature representation, and Fuzzy Inference Gait System-Deep Extreme Adaptive Fuzzy (FIGS-DEAF) based on Abide 1 dataset.

Lien vers le texte intégral (Open Access ou abonnement)