Article Sidebar

PDF
Issue: No. 11TA (2026)
Section: Article
DOI: 10.58490/ctjump.2026i11TA.4353
Date Published: 25/06/2026
Views 0
Downloads 0
How to Cite
Tran, T. N. P., Nguyen, V. T., Ayyakannu , E. A., Srinivasan, J. D., Vo, V. T., & Lu, T. D. (2026). APPLICATION OF ARTIFICIAL INTELLIGENCE IN SCREENING FOR AUTISM SPECTRUM DISORDER IN CHILDREN AT CAN THO CHILDREN’S HOSPITAL. Cantho Journal of Medicine and Pharmacy, 11TA, 86-92. https://doi.org/10.58490/ctjump.2026i11TA.4353

APPLICATION OF ARTIFICIAL INTELLIGENCE IN SCREENING FOR AUTISM SPECTRUM DISORDER IN CHILDREN AT CAN THO CHILDREN'S HOSPITAL

Thi Ngoc Phuong Tran1, , Van Trinh Nguyen1, Elavarasan Aishwarya Ayyakannu 1, Jyothi Dyana Srinivasan1, Van Thi Vo1, Tri Dien Lu1
1 Can Tho University of Medicine and Pharmacy

Main Article Content

Abstract

Background: Autism Spectrum Disorder is a complex neurodevelopmental disorder characterized by social-communication deficits and restricted, repetitive behaviors, with rising global prevalence, including in Vietnam, where early detection remains limited. Artificial intelligence (AI) provides a data-driven approach to improve diagnostic accuracy. Objectives: To develop an AI-based screening model for Autism Spectrum Disorder in children aged 18-36 months at Can Tho Children’s Hospital and to identify perinatal factors associated with Autism Spectrum Disorder. Materials and methods: A descriptive cross-sectional study was conducted on 980 children aged 18-36 months (November 2024-November 2025). Screening used M-CHAT-R/F, followed by DSM-5 confirmation. Logistic regression identified independent risk factors, and multiple AI models were evaluated using standard performance metrics. Results: Of 980 children, 43 (4.4%) were diagnosed with Autism Spectrum Disorder. Independent predictors included interventional delivery (aOR = 2.10), abnormal labor duration (aOR = 4.21), preterm birth (aOR = 2.31), and birth asphyxia (aOR = 7.24). M-CHAT-R/F Questions 1, 6, and 20 were most predictive. Decision Tree and Random Forest showed the best performance, with AUC values of approximately 0.80. Conclusions: Autism Spectrum Disorder prevalence was 4.4%, with interventional delivery, prolonged labor, preterm birth, and birth asphyxia as major perinatal risk factors. AI-based screening models demonstrated good discriminatory performance and may serve as useful tools to support early ASD screening in resource-limited healthcare settings.

Keywords

Autism spectrum disorder, Autism Spectrum Disorder diagnosis, Machine learning

Article Details

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

References

1. Bozkurt A, İpek H, Bayındır K, et al. Evaluation of prenatal and perinatal risk factors in autism spectrum disorder according to disease severity. Namik Kemal Med J. 2025. 13(1), 69-75, doi: 10.4274/nkmj.galenos.2025.80270.
2. Maenner MJ, Warren Z, Williams AR, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years - Autism and Developmental Disabilities Monitoring Network, 11 sites, United States, 2020. MMWR Surveill Summ. 2023. 72(2), 1-14, doi: 10.15585/mmwr.ss7202a1.
3. Thi Vui L, Duc DM, Thuy Quynh N, et al. Early screening and diagnosis of autism spectrum disorders in Vietnam: A population-based cross-sectional survey. J Public Health Res. 2022.11(2), 2460, doi: 10.4081/jphr.2021.2460.
4. Eigsti IM. The autism constellation and neurodiversity: Long-term and adult outcomes in autism spectrum disorder. Pediatr Clin North Am. 2024. 71(2), 327-341, doi: 10.1016/j.pcl.2024.01.003.
5. Mohammadi F, Shahrokhi H, Asadzadeh A, Pirmoradi S, Moghtader A, et al. Artificial intelligence in autism spectrum disorder diagnosis: A scoping review of face, voice, and text analysis methods. Health Sci Rep. 2025. 8(11), e71476, doi: 10.1002/hsr2.71476.
6. Shahini A, Kamath AP, Sharma E, Salvi M, Tan RS, Siuly S, et al. A systematic review for artificial intelligence-driven assistive technologies to support children with neurodevelopmental disorders. Inf Fusion. 2025. 124, 103441, doi: 10.1016/j.inffus.2025.103441.
7. Vo VT, Tran DT, Nguyen MP. Study on the prevalence of children from 18 to 36 months at risk of autism spectrum disorder by using M-CHAT-R/F scaleat Ca Mau province, Vietnam in 2022. Vietnam Medical Journal. 2023. 520(1B), doi: 10.51298/vmj.v520i1B.3905.
8. Crump C, Sundquist J, Sundquist K. Preterm or early term birth and risk of autism. Pediatrics. 2021.148(3),e2020032300, doi: 10.1542/peds.2020-032300.
9. Arun P, Azad C, Kaur G, Sharma P. A community-based study of antenatal and neonatal risk factors in autism spectrum disorder. Prim Care Companion CNS Disord. 2023. 25(2), 22m03339, doi: 10.4088/PCC.22m03339.
10. Silva VBD, Maia FA, Oliveira AJS, et al. Association between autism spectrum disorder and peripartum events: A case-control study. Rev Paul Pediatr. 2023. 41, e2021220, doi: 10.1590/1984-0462/2023/41/2021220.
11. Yuan JJ, Zhao YN, Lan XY, Zhang Y, Zhang R. Prenatal, perinatal and parental risk factors for autism spectrum disorder in China: a case-control study. BMC Psychiatry. 2024. 24(1), 219, doi: 10.1186/s12888-024-05643-0.
12. Preciado C, Baida M, Li Y, Li Y, Demopoulos C. Prenatal exposure to hypoxic risk conditions in autistic and neurotypical youth: Associated ventricular differences, sleep disturbance, and sensory processing. Autism Res. 2024. 17(12), 2547-2557, doi: 10.1002/aur.3250.
13. Zambre G, Albhar H, Masli T, Patil S. Detection and analysis of autism spectrum disorder using random forest classifier. In: Proceedings of the 2024 2nd International Conference on Recent Trends in Microelectronics, Automation, Computing and Communications Systems (ICMACC). IEEE. 2024, doi: 10.1109/ICMACC62921.2024.10894067.