Use of Machine Learning for Diagnosis in Students with Autism Spectrum Disorder Using Public Datasets

• Sara R. A. Leal University of Pernambuco (UPE)
• Suzerlly V. L. Pires University of Pernambuco (UPE)
• Vanderlene C. Rocha University of Pernambuco (UPE)
• Franciely A. de Souza University of Pernambuco (UPE)
• Lucas C. Teixeira University of Pernambuco (UPE)
• Joao F. L. de Oliveira University of Pernambuco (UPE)
• Ticia C. F. Cavalcante Federal University of Pernambuco (UFPE)
• Diego M. P. F. Silva University of Pernambuco (UPE)
• Carlo M. R. da Silva University of Pernambuco (UPE)

DOI: https://doi.org/10.5753/sbie.2024.241708

Abstract

Autism Spectrum Disorder (ASD) is a neurological condition that affects neurodevelopment, communication, and social interaction. It is often underreported, leading to educational challenges due to the lack of appropriate interventions. This study aims to develop a tool that assists educators in diagnosing ASD by utilizing Machine Learning algorithms to detect ASD signs across different ages, based on simple data extracted from three public datasets. These datasets were pre-processed and balanced using the SMOTE technique, and five algorithms were applied: Decision Tree, Random Forest, KNN, Naive Bayes, and Deep Learning. Random Forest stood out for its superior performance, with high accuracy and low error incidence. The results suggest that these models can be effective tools for early ASD screening, offering significant support to educators.

Keywords: Autism Spectrum Disorder, Machine Learning, Datasets

References

ARAÚJO, P. H.; DOS SANTOS, V. A.; BORGES, I. C. O autismo e a inclusão na educação infantil: estudo e revisão. Brazilian Journal of Development, 2021. DOI: 10.34117/bjdv7n2-563.

BRASIL. Ministério da Saúde. Diretrizes de Atenção à Reabilitação da Pessoa com Transtornos do Espectro do Autismo (TEA). Brasília: Ministério da Saúde, 2014.

BRENTANI, H. et al. Autism spectrum disorders: an overview on diagnosis and treatment. Brazilian Journal of Psychiatry, v. 35, p. S62–S72, 2013.

DENG, L.; RATTADILOK, P.; XIONG, R. A machine learning-based monitoring system for attention and stress detection for children with autism spectrum disorders. In: Proceedings of the International Conference on Intelligent Medicine and Health. ACM, 2021.

FROTA, M. et al. Aplicação de Árvore de Decisão para auxílio ao diagnóstico do transtorno do espectro autista. In: Anais da VII Escola Regional de Computação Aplicada à Saúde. SBC, 2019. Disponível em: [link]. Acesso em: 14 ago. 2023.

GIOIA, P. S. et al. Protocolo de avaliação e intervenção precoces de sinais de risco de autismo: comparando grupos de alto e baixo risco. SciELO Preprints, 2021.

GOIS, T. et al. Risk identification for autistic spectrum disorder in preschool children: design and validation of a screening instrument. SciELO Preprints, 2022.

GUPTA, K. N.; HAFIZ, G. Accurate estimate of autism spectrum disorder in children utilizing several machine learning techniques. In: 14th International Conference on Computational Intelligence and Communication Networks (CICN). IEEE, 2022.

GYORI, M. et al. Automated vs human recognition of emotional facial expressions of high-functioning children with autism in a diagnostic-technological context: Explorations via a bottom-up approach. In: Lecture Notes in Computer Science. Springer International Publishing, 2018, p. 466–473.

INEP. Caderno de conceitos e orientações do censo escolar 2021. Disponível em: [link]. Acesso em: 25 maio 2022.

LOWRI, C. Issues in persistent non-attendance at school of autistic pupils and recommendations following the reintegration of 11 autistic pupils. Good Autism Practice (GAP), v. 22, p. 12–20, 2021.

MUNKHAUGEN, E. et al. School refusal behaviour: are children and adolescents with autism spectrum disorder at a higher risk? Research in Autism Spectrum Disorders, v. 41-42, p. 31–38, 2017.

OVERLAND, E. et al. Exploring life with autism: Quality of life, daily functioning and compensatory strategies from childhood to emerging adulthood: A qualitative study protocol. Frontiers in Psychiatry, v. 13, 2022.

SANTOS, J. O. L. et al. O atendimento educacional especializado para os educandos com autismo na rede municipal de manaus-am. Revista Brasileira de Estudos Pedagógicos (RBEP), 2021. DOI: 10.24109/2176-6681.rbep.102.i260.4150.

SHINDE, A. V.; PATIL, D. D. A multi-classifier-based recommender system for early autism spectrum disorder detection using machine learning. Healthcare Analytics, v. 4, p. 100211, 2023.

SHOHIEB, S. M. et al. Early detection of autism by extracting features: A case study in Bangladesh. In: International Conference on Robotics, Electrical and Signal Processing Techniques (ICREST). IEEE, 2019.

WANG, H. et al. Social skills assessment in young children with autism: A comparison evaluation of the SSRS and PKBS. Journal of Autism and Developmental Disorders, v. 41, n. 11, p. 1487–1495, 2011.

YANG, X.; ISLAM, M. S.; KHALED, A. M. A. Functional connectivity magnetic resonance imaging classification of autism spectrum disorder using the multisite ABIDE dataset. In: IEEE EMBS International Conference on Biomedical & Health Informatics (BHI). IEEE, 2019.

YESILYURT, T. H.; DIAGNOSING, S. Diagnosing autism spectrum disorder using machine learning techniques. In: 6th International Conference on Computer Science and Engineering (UBMK). IEEE, 2021.