International Journal of Advanced Culture Technology

The International Promotion Agency of Culture Technology (국제문화기술진흥원)
권호 표지
DOI QR코드

DOI QR Code

Utilizing Deep Learning for Early Diagnosis of Autism: Detecting Self-Stimulatory Behavior

  • Seongwoo Park (Department of Global Entrepreneurship and Information Communication Technology, Handong Global University) ;
  • Sukbeom Chang (Department of Global Entrepreneurship and Information Communication Technology, Handong Global University) ;
  • JooHee Oh (Department of Management and Economics, Handong Global University)
  • Received : 2024.07.25
  • Accepted : 2024.09.05
  • Published : 2024.09.30
Download PDF
⟨ Previous Next ⟩

Abstract

We investigate Autism Spectrum Disorder (ASD), which is typified by deficits in social interaction, repetitive behaviors, limited vocabulary, and cognitive delays. Traditional diagnostic methodologies, reliant on expert evaluations, frequently result in deferred detection and intervention, particularly in South Korea, where there is a dearth of qualified professionals and limited public awareness. In this study, we employ advanced deep learning algorithms to enhance early ASD screening through automated video analysis. Utilizing architectures such as Convolutional Long Short-Term Memory (ConvLSTM), Long-term Recurrent Convolutional Network (LRCN), and Convolutional Neural Networks with Gated Recurrent Units (CNN+GRU), we analyze video data from platforms like YouTube and TikTok to identify stereotypic behaviors (arm flapping, head banging, spinning). Our results indicate that the LRCN model exhibited superior performance with 79.61% accuracy on the augmented platform video dataset and 79.37% on the original SSBD dataset. The ConvLSTM and CNN+GRU models also achieved higher accuracy than the original SSBD dataset. Through this research, we underscore AI's potential in early ASD detection by automating the identification of stereotypic behaviors, thereby enabling timely intervention. We also emphasize the significance of utilizing expanded datasets from social media platform videos in augmenting model accuracy and robustness, thus paving the way for more accessible diagnostic methods.

Keywords

References

  1. F. Negin, B. Ozyer, S. Agahian, S. Kacdioglu, and G. T. Ozyer, "Vision-assisted recognition of stereotype behaviors for early diagnosis of autism spectrum disorders," Neurocomputing, vol. 446, pp. 145-155, 2021. 
  2. H. Alkahtani, Z. A. Ahmed, T. H. Aldhyani, M. E. Jadhav, and A. A. Alqarni, "Deep learning algorithms for behavioral analysis in diagnosing neurodevelopmental disorders," Mathematics, vol. 11, no. 19, p. 4208, 2023. 
  3. E. Fernell, M. A. Eriksson, and C. Gillberg, "Early diagnosis of autism and impact on prognosis: A narrative review," Clinical Epidemiology, pp. 33-43, 2013. 
  4. J. I. Kim, and H. J. Yoo, "Diagnosis and assessment of autism spectrum disorder in South Korea," Journal of the Korean Academy of Child and Adolescent Psychiatry, vol. 35, no. 1, p. 15, 2024. 
  5. S. Lee, and J. Moon, "A review of early screening for autism spectrum disorder," Journal of Psychological Movement, vol. 8, no. 1, pp. 55-79, 2022. 
  6. C. Yoon, "Standardization of the Korean-Autism Diagnostic Scale and development of a web-based evaluation system," Journal of Emotional and Behavioral Disorders, vol. 20, no. 3, pp. 27-43, 2004. 
  7. J. Choi, J. Park, J. Kim, and M. Kim, "Validation study of early diagnosis tools for autism spectrum disorder based on smart platform," Journal of Autism and Developmental Disorders, vol. 22, no. 1, pp. 137-156, 2022. 
  8. H. Sunwoo, D. Noh, K. Kim, J. Kim, and H. Yoo, "Perceptions of professionals related to early screening for autism spectrum disorder," Journal of Child and Adolescent Psychiatry, vol. 28, no. 2, pp. 96-105, 2017. 
  9. I. Lovaas, C. Newsom, and C. Hickman, "Self-stimulatory behavior and perceptual reinforcement," Journal of Applied Behavior Analysis, vol. 20, no. 1, pp. 45-68, 1987. 
  10. P. Wei, D. Ahmedt-Aristizabal, H. Gammulle, S. Denman, and M. A. Armin, "Vision-based activity recognition in children with autism-related behaviors," Heliyon, vol. 9, no. 6, p. e12345, 2023. 
  11. D. L. Robins, D. Fein, M. L. Barton, and J. A. Green, "The Modified Checklist for Autism in Toddlers: An initial study investigating the early detection of autism and pervasive developmental disorders," Journal of Autism and Developmental Disorders, vol. 31, pp. 131-144, 2001. 
  12. A. Anzulewicz, K. Sobota, and J. T. Delafield-Butt, "Toward the autism motor signature: Gesture patterns during smart tablet gameplay identify children with autism," Scientific Reports, vol. 6, no. 1, p. 31107, 2016. 
  13. A. Lakkapragada, A. Kline, O. C. Mutlu, K. Paskov, B. Chrisman, N. Stockham, et al., "The classification of abnormal hand movement to aid in autism detection: Machine learning study," JMIR Biomedical Engineering, vol. 7, no. 1, p. e33771, 2022. 
  14. S. Rajagopalan, A. Dhall, and R. Goecke, "Self-stimulatory behaviours in the wild for autism diagnosis," in Proceedings of the IEEE International Conference on Computer Vision Workshops, pp. 755-761, 2013. 
  15. X. Shi, Z. Chen, H. Wang, D. Y. Yeung, W. K. Wong, and W. C. Woo, "Convolutional LSTM network: A machine learning approach for precipitation nowcasting," in Advances in Neural Information Processing Systems, vol. 28, 2015. 
  16. J. Donahue, L. Anne Hendricks, S. Guadarrama, M. Rohrbach, S. Venugopalan, K. Saenko, and T. Darrell, "Long-term recurrent convolutional networks for visual recognition and description," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2625-2634, 2015.