1. Almulla AA, Khasawneh MAS. Assessing AI-Based Large Language Models (ChatGPT, Google Gemini, and DeepSeek) for Common Parent Questions about Autism: Acceptability, Readability, and Accuracy. Psychiatr Q. 2025.

With growing reliance on AI chatbots for parenting support, this study presents the first evaluation of large language models (LLMs) in addressing common autism-related questions. It compared ChatGPT, Google Gemini, and DeepSeek based on the accuracy, clarity, and usefulness of their responses. The findings aim to inform parents and clinicians about the strengths and limitations of using AI tools in early ASD care. Twenty common questions about Autism Spectrum Disorder (ASD) were identified through content analysis of social media, Google Trends, and ASD forums. These questions were refined by two educational psychologists, and standardized benchmark answers were created by a panel of pediatric neurodevelopment specialists. Two blinded pediatric autism experts then evaluated the AI-generated responses based on quality, as well as usefulness and reliability. GPT-4 achieved the highest mean quality score (M = 4.85, SD = 0.36), followed by Gemini and DeepSeek (both M = 4.55, SD = 0.51; p > 0.05). For usefulness, GPT-4 scored M = 6.40 (SD = 0.75), Gemini M = 6.10 (SD = 0.85), and DeepSeek M = 6.05 (SD = 0.82; p > 0.05). In reliability ratings, Gemini led with M = 6.40 (SD = 0.82), GPT-4 M = 6.25 (SD = 0.71), and DeepSeek M = 5.95 (SD = 0.94; p > 0.05). Findings indicated that AI-based chatbots, by providing rapid, comprehensible, and evidence-based guidance on early signs, interventions, and family support, demonstrate significant potential in bridging the information gap for parents-especially when access to specialists is limited.

Lien vers le texte intégral (Open Access ou abonnement)

2. Anderle F, Pasqualotto A, Bentenuto A, Venuti P, Benso F. Assessing executive attention in autistic children: strengths, weaknesses and individual differences. Child Neuropsychol. 2025: 1-23.

Research consistently shows that autistic children often exhibit cognitive challenges, particularly in executive functions (EFs), since the preschool years. EFs are cognitive abilities that help regulate impulses, manage information, filter distractions, and shift focus between tasks. Various performance-based measures have been developed to assess EFs in autistic children. However, inconsistencies in findings have raised concerns about the ability of traditional EF measures to capture the real-life challenges these children face, largely due to reductionist approaches and the overlooked issue of task impurity. Here, we employed a broader comprehensive battery – the Measures of Executive Attention – to assess EFs in 43 autistic children aged 8-14 years, compared to 43 neurotypicals matched for age, sex, and fluid reasoning index. The results revealed that neurotypicals outperformed autistic children in most tasks. Specifically, autistic children showed lower performance in cognitive flexibility and generative thinking in a graphical task, as well as in working memory capacity under novel and emotionally stressful conditions. However, when fine motor and verbal skills were excluded from the composite score calculation, no group differences emerged in certain tasks, such as visual search and working memory capacity in a familiar exercise. Our findings highlight the importance of assessing executive attention through multidimensional and context-sensitive tools and offer new insights into cognitive variability in autism.

Lien vers le texte intégral (Open Access ou abonnement)

3. Choi TY, Gunawan A, Seo D, Park J, Ahn EH, Suh SW, Fuccillo MV, Choi K. Applying biologically anchored subtypes to advance precision medicine in autism spectrum disorder. Neurobiol Dis. 2025: 107219.

Autism spectrum disorder (ASD) is heterogeneous at every level, from behavior to molecular pathways, limiting the value of subgrouping schemes built on surface phenotypes alone. We synthesize evidence that biologically anchored subtypes, defined by convergent genetics, developmental timing, and brain-body crosstalk, offer a tractable path to precision medicine. Leveraging advances in large-scale genomic resources and computational analytics, we propose a multi-axis framework: (i) genetic architecture spanning rare variants and polygenic load, (ii) developmental windows from mid-gestation to infancy divergence and regression, and (iii) brain-body interactions shaping plasticity and symptom expression. This framework enables mechanism-guided therapeutic strategies through biomarker-stratified enrollment, target-engagement readouts, and circuit-anchored outcomes. Preclinical platforms, genetically engineered mice and patient-derived induced pluripotent stem cells (iPSCs), demonstrate convergence onto limited synaptic and connectivity « neurotypes, » enabling causal links from gene to circuit to behavior and proof-of-concept rescue. We close with priorities: standardized multi-platform characterization, decision tools linking subtype labels to interventions, and stratified trials that co-report clinical and biological endpoints, with ethical guardrails to ensure early stratification expands opportunity while advancing individualized care.

Lien vers le texte intégral (Open Access ou abonnement)

4. Fadl-Elmula I, Abdel-Raheem SY, Khalid R. Atypical case of Rett syndrome with concurrent MECP2 gene mutation and del(15)(q22qter) karyotype: A case report and review of literature. World J Clin Pediatr. 2025; 14(4): 109874.

BACKGROUND: Rett syndrome is a monogenic X-linked dominant condition that affects 1/(10000-15000) girls due to de novo mutations in the methyl-CpG binding protein 2 (MECP2) gene mapped to chromosome Xq28. The disease-causing gene was identified as a mutation in the MECP2 gene, which is found in approximately 80% of patients diagnosed with Rett syndrome. Although chromosomal changes resulting in del(15)(q11q13) are usually associated with Angelman and Prader-Willi syndrome, very few cases, if any, of Rett syndrome with terminal 15q22-qter deletion have been published in English literature. CASE SUMMARY: In this study, we report an unusual and rare clinical presentation of Rett syndrome in a 12-year-old Sudanese girl. The patient was brought in by her parents, complaining of gradual onset of abnormal walking, abnormal hand movement, loss of speech, and mental retardation for ten years. There was no reported history of convulsions or loss of consciousness. Clinical examination revealed microcephaly with no other apparent dysmorphic features, intact cranial nerves, and abnormal gait. She showed repetitive and stereotyped behaviors, including hand flapping, stimming, and chest pounding, which were concomitant with autism spectrum disorder. Magnetic resonance imaging and electroencephalography investigations were normal, and cytogenetic analysis showed 46,XX, del(15)(q22qter). Further molecular analysis using whole sequencing of MECP2 revealed an alteration cytosine > thymine at nucleotide 401, leading to phenylalanine replacing a serine at amino acid position 134. CONCLUSION: This case, the first reported instance of Rett syndrome in Sudan, is of significant interest. The patient carries both the MECP2 gene mutation and the chromosome 15q22-qter deletion, which may explain the autistic behavior with atypical presentation of Rett syndrome. This report expands the genetic diversity of Rett syndrome, demonstrating how co-occurring 15q22-qter deletions can reshape MECP2-associated phenotypes in Rett syndrome.

Lien vers le texte intégral (Open Access ou abonnement)

5. Ferguson EF, Spackman E, Hardan AY, Uljarević M. Self-injurious behaviors with increased likelihood of injury in autistic youth: The role of distress linked to a strong preference for sameness. Autism. 2025: 13623613251396036.

Self-injurious behaviors in autistic youth vary widely in their form, intensity, and potential for physical injury. This study explored predictors of two categories of self-injurious behaviors that were delineated based on their potential for physical injury (self-injurious behavior-High Likelihood and self-injurious behavior-Low Likelihood), with a focus on the role of distress linked to a strong preference for routines (otherwise known as insistence on sameness). The sample included 1892 autistic youth (M(age) = 10.82, SD(age) = 4.14; 22% females) across the United States. Distress associated with insistence on sameness was the strongest predictor of total self-injurious behavior scores, self-injurious behavior-Low Likelihood, and self-injurious behavior-High Likelihood, after controlling for demographic factors and speech level, and remained a unique predictor after accounting for variance explained by other restricted and repetitive behaviors. Sensory hypersensitivity and sensory-seeking were strong unique predictors of all self-injurious behaviors, while hyposensitivity was a weaker predictor of self-injurious behavior total and self-injurious behavior-Low Likelihood, and a non-significant predictor of self-injurious behavior-High Likelihood. Among demographic factors, lower household income was the strongest predictor of all self-injurious behaviors. Higher speech level was a positive predictor of self-injurious behavior-Low Likelihood but a negative predictor of self-injurious behavior-High Likelihood. These findings demonstrate the role of distress associated with insistence on sameness in manifestations of self-injurious behaviors and highlight the importance of exploring predictors at a more granular level to inform targeted interventions and support.Lay AbstractThis study explored factors associated with repetitive self-injurious behaviors in autistic youth, focusing on emotional distress linked to a strong preference for routines (often referred to as insistence on sameness), a category of behaviors that includes difficulties with change and a strong preference for routines and/or rituals. We examined two categories of self-injurious behaviors: lower likelihood of physical injury (self-injurious behavior-Low Likelihood) and higher likelihood of physical injury (self-injurious behavior-High Likelihood). The study included 1892 autistic youth of varying ages and cognitive abilities. The main finding was that distress associated with disruptions to routines was the strongest predictor of all types of self-injurious behavior. Youth with greater sensitivity to sensory input or sensory-seeking behaviors were more likely to engage in self-injurious behaviors that could increase risk of physical injury. In contrast, those with reduced sensitivity to sensory input were less likely to engage in self-injurious behaviors. Demographic factors also played a role in the manifestation and severity of self-injurious behaviors. Lower household income was strongly associated with greater self-injurious behavior severity. In addition, higher speech production (e.g. speaking in full sentences) was associated with greater severity of self-injurious behavior-Low Likelihood and self-injurious behavior-High Likelihood. These findings highlight the complexity of self-injurious behaviors in autistic youth and the importance of understanding the different factors that contribute to these behaviors. This study may help to contribute to the development of more responsive, tailored interventions for self-injurious behaviors among autistic youth.

Lien vers le texte intégral (Open Access ou abonnement)

6. Güngör İ, Sarman A, Tuncay S. Can artificial intelligence and face recognition using deep learning detect emotions in children with autism?. PLoS One. 2025; 20(12): e0338701.

BACKGROUND/OBJECTIVES: This study aimed to evaluate the performance of deep learning models for recognizing facial expressions of children with autism through face recognition technologies. METHODS: Conducted between November 2024 and February 2025, this research introduced a novel deep learning architecture, AutismEfficientNet, by integrating EfficientNetV2 and MobileNetV3. Two previously published datasets-Dr. Fatma M. Talaat’s Emotions of Autistic Children and the FERAC dataset-were used, both containing labeled facial images of children with autism displaying various emotions. Images were preprocessed and divided into training, validation, and test sets. Emotion classification was performed using convolutional neural networks. Models were evaluated using accuracy, sensitivity, specificity, and F1 scores. RESULTS: Deep learning models were able to accurately recognize emotional expressions in children with autism. The proposed AutismEfficientNet achieved the highest accuracy (75.8%) among the tested models, outperforming EfficientNetV2-L (72.4%) and MobileNetV3-L (70.2%). CONCLUSIONS: The findings indicate that AutismEfficientNet provides improved classification accuracy compared to standard architectures when applied to datasets of children with autism. Further validation is required in clinical settings and diverse populations before integration into pediatric care.

Lien vers le texte intégral (Open Access ou abonnement)

7. Hasan H, Hagerman R, Say DS, Nguyen AP, Babata K, Oyegbile-Chidi T, Herrera-Guerra A, Torrents C, Silver CE, Restrepo B. Parallel paths: A narrative review exploring autism and its co-occurring conditions. World J Clin Pediatr. 2025; 14(4): 111641.

Autism is a heterogeneous condition with a rising prevalence and demand for specialized care. Autistic children are more likely than neurotypical peers to experience co-occurring conditions (CCs), including medical, psychiatric, and behavioral issues, highlighting the urgent need for autism-competent healthcare providers in general healthcare. This review aims to equip primary care providers (PCPs) with a concise summary of common CCs and strategies for effective identification. A panel of experts with extensive experience in caring for autistic children collaboratively summarized key literature, research evidence, and existing clinical trial outcomes, supplementing their clinical expertise. Autistic children consistently show higher rates of both medical and mental health issues. Despite greater healthcare utilization, many autistic individuals report unmet needs. CCs can impair behavior, functioning, and well-being, but are often treatable when recognized early. Timely identification and management of medical and psychiatric CCs are critical for improving outcomes for autistic children and their families. This evidence-based review supports PCPs in enhancing their knowledge, fostering early recognition, and delivering comprehensive, responsive care.

Lien vers le texte intégral (Open Access ou abonnement)

8. Monteiro DC, de Goes Omena CP, Lira E, Lins V. Successful Electroconvulsive Therapy for Severe Treatment-resistant Depression in a Young Woman With Autism, Attention-deficit/Hyperactivity Disorder, and Cardiac Pacemaker: A Case Report. J ect. 2025.

Lien vers le texte intégral (Open Access ou abonnement)

9. Prasad S, Travis L, Mei C, Younis R. Understanding Complications of Sinusitis in Paediatric Patients With Autism: A KID Database Study. Clin Otolaryngol. 2025.

OBJECTIVE: This study evaluated the risk of orbital and neurological complications of acute sinusitis in pediatric patients with autism spectrum disorder (ASD) compared to neurotypical peers. STUDY DESIGN: Retrospective cohort study. SETTING: Analysis of data from the 2019 Kids’ Inpatient Database (KID), a nationally representative sample of pediatric inpatient hospitalizations. METHODS: Paediatric patients discharged with acute sinusitis were identified using ICD-10-CM codes and stratified by the presence of an ASD diagnosis. Primary outcomes included orbital complications (e.g., eye pain, chemosis, cellulitis, abscess) and neurological complications (e.g., seizures, meningitis, stroke). Secondary outcomes included surgical interventions, length of stay (LOS), and hospital costs. One-to-one propensity score matching (PSM) was performed to adjust for patient- and hospital-level factors. RESULTS: Of 3763 hospitalised children with acute sinusitis, 125 (3.4%) had ASD. After matching, ASD patients had significantly higher rates of seizures (26.4% vs. 6.3%; OR = 3.52; p < 0.001) but were less likely to undergo maxillary sinus surgery (OR = 0.46; p = 0.037). Orbital complications, including eye pain, were more frequent in the ASD group (4% vs. 1.4%; p = 0.043). LOS and hospital costs did not differ significantly between groups after matching. CONCLUSION: Children with ASD hospitalized for acute sinusitis appear to be at higher risk for seizures but less likely to undergo certain sinus surgeries. These findings underscore the need for tailored neurological management and exploration of factors influencing surgical decisions in this population.

Lien vers le texte intégral (Open Access ou abonnement)

10. Rosen BF, Sumaria K, Miller CE, Arbuckle C, McNamara T, Callaghan T, Mazhani T, Rowley L, Rivalta-Dallal A, English K, Foulkes E, Julies P. ‘See me Autism’: improving the experience and safety of autistic children in the paediatric emergency department. Arch Dis Child Educ Pract Ed. 2025.

Implementation of co-produced training and resources can improve the experience and safety of autistic children in a paediatric emergency department.

Lien vers le texte intégral (Open Access ou abonnement)

11. Suspitsin EN, Malysheva KS, Laptiev SA, Sharonova OS, Abuzova AS, Kuznitsyna AA, Melashenko TV, Efremova OV, Korzun PR, Binnatova JO, Gorgul YA, Syomina MV, Imyanitov EN. Monogenic defects in Russian children with autism spectrum disorders. World J Clin Pediatr. 2025; 14(4): 108733.

BACKGROUND: Autism spectrum disorders (ASD) represent a substantial social problem affecting at least 1 in 100 children worldwide. These conditions are very often accompanied by intellectual disability (ID) and speech delay; thus, they can be considered within a clinical continuum of neurodevelopmental disorders. Given the high heterogeneity of ASD, the subjective nature of diagnostic criteria, and the presence of phenocopies, identifying genetic determinants of these disorders remains a challenge. AIM: To investigate the spectrum and frequency of rare genetic variants in genes with proven association with ASD in Russian children. METHODS: 110 patients from 106 families were recruited into the study (mean age at diagnosis 6 years; boy-to-girl ratio 3:1. Most of the patients (84%) demonstrated a combination of ASD with developmental delay (DD) or ID. Patients with syndromic features were subjected to the chromosomal microarray analysis. The remained children underwent clinical exome sequencing aimed at identifying presumably monogenic causes of ASD. The study focused on rare (minor allele frequency ≤ 0.001) variants affecting high-confidence ASD-associated genes. RESULTS: Pathogenic copy number variations were detected in three (7%) of the patients examined. Clinical exome sequencing revealed pathogenic/likely pathogenic variants in 12 of 105 cases (11%), indicating the presence of monogenic syndromes with established clinical significance (Pitt-Hopkins syndrome, ZTTK syndrome, syndromic X-linked ID of Billuart type, Snijders-Blok-Campeau, Helsmoortel-van der Aa, Coffin-Siris, Clark-Baraitser, Keefstra syndromes, etc.). In addition, 27 patients (26%) had 37 rare variants of unknown clinical significance in DSCAM, SHANK2, AUTS2, ADNP, ANKRD11, APBA2, ARID1B, ASTN2, ATRX, SCN1A, CHD2, DEAF1, EHMT1, GRIN2B, NBEA, NR4A2, TRIO, TRIP12, POGZ, EP300, FOXP1, PCDH19, GRIN2A, NCKAP1, and CHD8 genes. No specific variant was detected more than once in unrelated patients. Among the genes with rare variants found in 2 or more patients were TRIP12 (n = 4), AUTS2 (n = 3), ARID1B (n = 3), PCDH19 (n = 3), EP300 (n = 3), TRIO (n = 2), ASTN2 (n = 2), EHMT1 (n = 2), and CHD2 (n = 2). Of note, 5 male ASD/DD patients from 3 unrelated families had PCDH19 missense variants, confirming that at least some hemizygous males with non-mosaic PCDH19 variants may present with neurobehavioral abnormalities. These variants did not cause epilepsy restricted to females in patients’ mothers or sisters. CONCLUSION: These data confirm a tremendous diversity of genetic causes of ASD. Clinical exome sequencing may serve as a reasonable alternative to whole-exome sequencing.

Lien vers le texte intégral (Open Access ou abonnement)