Artificial intelligence for detecting autism spectrum disorder: a systematic review and meta-analysis
- 1Department of Pediatrics, West China Second University Hospital, Sichuan University
- 2Key Laboratory of Birth Defects and Related Diseases of Women and Children (Sichuan University), Ministry of Education
Abstract
Background: Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental condition, and early identification is essential for timely intervention. Conventional diagnosis relies on clinical interviews, behavioral observations, caregiver reports, and standardized tools, which are time-consuming and resource-dependent. Artificial intelligence (AI) offers opportunities for screening and adjunctive diagnosis, but studies vary in data sources, model types, and validation methods, requiring systematic evaluation.
Methods: We searched PubMed, Embase, the Cochrane Library, Web of Science, and IEEE Xplore for studies of AI-based ASD detection. Eligible studies were required to provide ASD-versus-non-ASD diagnostic-accuracy data that could be extracted or reliably reconstructed at the independent participant or case level. Risk of bias and applicability were assessed using a modified QUADAS-AI tool. Pooled sensitivity and specificity were estimated with a bivariate Reitsma random-effects model. HSROC AUCs were calculated with bias-corrected and accelerated (BCa) 95% confidence intervals based on 5,000 study-level bootstrap replicates and were omitted when SROC fits were unstable. Sensitivity analyses addressed dataset overlap, model selection, and reference-standard quality.
Results: Of 17,387 records, 26 studies were included. Pooled sensitivity was 0.908 (95% CI: 0.861–0.940), pooled specificity was 0.868 (95% CI: 0.830–0.898), and the HSROC AUC was 0.936 (BCa 95% CI: 0.860–0.973). A dataset-independent analysis of 21 studies yielded lower estimates (sensitivity 0.870, specificity 0.841, AUC 0.900). Leave-one-out results were generally stable. Funnel-plot asymmetry was not significant in the primary analysis and varied in direction and statistical significance across sensitivity analyses.
Conclusions: AI-based models showed relatively high pooled diagnostic performance for ASD detection, but substantial heterogeneity, methodological and applicability concerns, limited external validation, and potential non-independence of underlying datasets limit the certainty and generalizability of the evidence. Funnel-plot analyses did not provide consistent evidence of conventional small-study effects or selective reporting. AI should therefore be considered an adjunct rather than a standalone diagnostic tool. Prospective multicenter validation in representative populations, participant-level evaluation, clearly defined reference diagnoses, prespecified models and operating thresholds, and independent evaluation datasets are needed before broader clinical implementation.
Related
- Notification: Artificial Intelligence in Autism Spectrum Disorder Diagnosis: A Scoping Review of Face, Voice, and Text Analysis Methods.
- Psychopharmacological Effects of Cannabidiol in Rodent Models of Autism Spectrum Disorder: A Systematic Review of Behavioral Outcomes, Neurobiological Mechanisms, and Methodological Quality
- Mapping the interplay of intelligence, empathy, systemizing, and autistic traits in children with and without autism spectrum disorder.
- Early Detection of Red Palm Weevil Infestation: A Systematic Review of Sensing Technologies, Artificial Intelligence, and Field Readiness
- Cochlear Implants Versus Hearing Aids in Children With Auditory Neuropathy Spectrum Disorder: A Systematic Review
BibTeX
@article{wang2026artificial,
title = {{Artificial intelligence for detecting autism spectrum disorder: a systematic review and meta-analysis}},
author = {Qian Wang and Rong Luo and Qianhui Wen},
journal = {Frontiers in Public Health},
year = {2026},
month = oct,
doi = {10.3389/fpubh.2026.1923851},
url = {https://www.frontiersin.org/articles/10.3389/fpubh.2026.1923851/full},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Artificial intelligence for detecting autism spectrum disorder: a systematic review and meta-analysis
AU - Qian Wang
AU - Rong Luo
AU - Qianhui Wen
PY - 2026
DA - 2026/10/12
JO - Frontiers in Public Health
DO - 10.3389/fpubh.2026.1923851
UR - https://www.frontiersin.org/articles/10.3389/fpubh.2026.1923851/full
AB - Background: Autism spectrum disorder (ASD) is a heterogeneous neurodevelopmental condition, and early identification is essential for timely intervention. Conventional diagnosis relies on clinical interviews, behavioral observations, caregiver reports, and standardized tools, which are time-consuming and resource-dependent. Artificial intelligence (AI) offers opportunities for screening and adjunctive diagnosis, but studies vary in data sources, model types, and validation methods, requiring systematic evaluation. Methods: We searched PubMed, Embase, the Cochrane Library, Web of Science, and IEEE Xplore for studies of AI-based ASD detection. Eligible studies were required to provide ASD-versus-non-ASD diagnostic-accuracy data that could be extracted or reliably reconstructed at the independent participant or case level. Risk of bias and applicability were assessed using a modified QUADAS-AI tool. Pooled sensitivity and specificity were estimated with a bivariate Reitsma random-effects model. HSROC AUCs were calculated with bias-corrected and accelerated (BCa) 95% confidence intervals based on 5,000 study-level bootstrap replicates and were omitted when SROC fits were unstable. Sensitivity analyses addressed dataset overlap, model selection, and reference-standard quality. Results: Of 17,387 records, 26 studies were included. Pooled sensitivity was 0.908 (95% CI: 0.861–0.940), pooled specificity was 0.868 (95% CI: 0.830–0.898), and the HSROC AUC was 0.936 (BCa 95% CI: 0.860–0.973). A dataset-independent analysis of 21 studies yielded lower estimates (sensitivity 0.870, specificity 0.841, AUC 0.900). Leave-one-out results were generally stable. Funnel-plot asymmetry was not significant in the primary analysis and varied in direction and statistical significance across sensitivity analyses. Conclusions: AI-based models showed relatively high pooled diagnostic performance for ASD detection, but substantial heterogeneity, methodological and applicability concerns, limited external validation, and potential non-independence of underlying datasets limit the certainty and generalizability of the evidence. Funnel-plot analyses did not provide consistent evidence of conventional small-study effects or selective reporting. AI should therefore be considered an adjunct rather than a standalone diagnostic tool. Prospective multicenter validation in representative populations, participant-level evaluation, clearly defined reference diagnoses, prespecified models and operating thresholds, and independent evaluation datasets are needed before broader clinical implementation.
SN - 2296-2565
ER -