Autism Spectrum Disorder Equipped with Convolutional-cum-visual Attention Mechanism

PDF (686KB), PP.39-51

Views: 0 Downloads: 0

Author(s)

Ayesha Shaik 1,* Lavish R. Jain 2 Balasundaram A. 1

1. Center for Cyber Physical Systems, Vellore Institute of Technology, Chennai, 600127, India

2. School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, 600127, India

* Corresponding author.

DOI: https://doi.org/10.5815/ijitcs.2025.05.04

Received: 20 May 2025 / Revised: 8 Jul. 2025 / Accepted: 22 Aug. 2025 / Published: 8 Oct. 2025

Index Terms

Explainable AI, Autism Spectrum Disorder, Convolutional Neural Network, Visual Attention, Deep Learning

Abstract

This research work aims to utilize deep learning techniques to identify autism traits in children based on their facial features. By combining traditional convolutional approaches with attention layers, the study seeks to enhance interpretability and accuracy in identifying autism spectrum disorder (ASD) traits. The dataset includes diverse facial images of children diagnosed with ASD and neuro-typical children, ensuring comprehensive representation. Preprocessing techniques standardize and enhance image quality, mitigating biases. Integration of attention layers within the convolutional neural net-work (CNN) architecture focuses on crucial facial features, improving feature extraction and classification accuracy. This approach enhances model interpretability through eXplainable AI (XAI) techniques. Model training involves optimization and validation processes, employing hyper parameter tuning and cross-validation for robustness. The performance of this combined model yielded close to 95% accuracy outperforming existing models in terms of complexity to accuracy ratio.

Cite This Paper

Ayesha Shaik, Lavish R. Jain, Balasundaram A., "Autism Spectrum Disorder Equipped with Convolutional-cum-visual Attention Mechanism", International Journal of Information Technology and Computer Science(IJITCS), Vol.17, No.5, pp.39-51, 2025. DOI:10.5815/ijitcs.2025.05.04

Reference

[1]M. Mishra, and U. C. Pati, “Autism spectrum disorder detection using surface morphometric feature of smri in machine learning”, In 2021 8th international conference on smart computing and communications (icscc), IEEE, pp. 17-20, 2021.
[2]G. Ramaswami, and D. H. Geschwind, “Genetics of autism spectrum disorder”, Handbook of clinical neurology, vol. 147, pp. 321-329, 2018.
[3]C. Lord, M. Elsabbagh, G. Baird, and Veenstra- J. Vanderweele, “Autism spectrum disorder”, The lancet, vol. 392, no. 10146, pp. 508-520, 2018.
[4]M. Mergl, and C. A. S. Azoni, “Echolalia's types in children with Autism Spectrum Disorder”, Revista Cefac, vol. 17, pp. 2072-2080, 2015.
[5]D. K. Anderson, M. P. Maye, and C. Lord, “Changes in maladaptive behaviors from midchildhood to young adult-hood in autism spectrum disorder”, American journal on intellectual and developmental disabilities, vol. 116, no. 5, pp. 381-397, 2011.
[6]L. Zwaigenbaum, J. A. Brian, and A. Ip, “Early detection for autism spectrum disorder in young children”, Paediatrics & Child Health, vol. 24, no. 7, pp. 424-432, 2019.
[7]H. Hodges, C. Fealko, and N. Soares, “Autism spectrum disorder: definition, epidemiology, causes, and clinical evaluation”, Translational pediatrics, vol. 9, no. 1, pp. 55, 2020.
[8]N. BalaKrishna, M. M. Krishnan, S. M. Reddy, S. K. Irfan, and S. Sumaiya, “Autism spectrum disorder detection using machine learning” In 2023 3rd International Conference on Advance Computing and Innovative Technologies in Engineering (ICACITE), IEEE, pp. 1645-1650, 2023.
[9]J. Kang, X. Han, J. Song, Z. Niu, and X. Li, “The identification of children with autism spectrum disorder by SVM approach on EEG and eye-tracking data”, Computers in biology and medicine, vol. 120, pp. 103722, 2020.
[10]Z. A. Huang, Z. Zhu, C. H. Yau, and K. C. Tan, “Identifying autism spectrum disorder from resting-state fMRI using deep belief network”, IEEE Transactions on neural net-works and learning systems, vol. 32, no. 7, pp. 2847-2861, 2020.
[11]J. Zhang, F. Feng, T. Han, X. Gong, and F. Duan, “Detection of autism spectrum disorder using fMRI functional connectivity with feature selection and deep learning”, Cognitive Computation, vol. 15, no. 4, pp. 1106-1117, 2023.
[12]A. S. Heinsfeld, A. R. Franco, R. C. Craddock, A. Buchweitz, and F. Meneguzzi, “Identification of autism spectrum dis-order using deep learning and the ABIDE dataset”, Neu-roImage: Clinical, vol. 17, pp. 16-23, 2018
[13]M. S. Alam, M. M. Rashid, R. Roy, A. R. Faizabadi, K. D. Gupta, and M. M. Ahsan, “Empirical study of autism spec-trum disorder diagnosis using facial images by improved transfer learning approach”, Bioengineering, vol. 9, no. 11, 710, 2022.
[14]H. B. Asif Mohamed, S. Khan, and M. KG, “Early Detection of Autism Spectrum Disorder Using Enhanced Deep-Convolutional Neural Network (EdCnn) Models”, Journal of Pharmaceutical Negative Results, vol.13, pp. 7801-7805, 2022.
[15]A. Kanchana, and R. Khilar, “Prediction of autism spectrum disorder using random forest classifier in adults”, In 2022 IEEE 4th International Conference on Cybernetics, Cognition and Machine Learning Applications (ICCCMLA), IEEE, pp. 242-249, 2022.
[16]K. S. Omar, P. Mondal, N. S. Khan, M. R. K. Rizvi, and M. N. Islam, “A machine learning approach to predict autism spectrum disorder”, In 2019 International conference on electrical, computer and communication engineering (EC-CE), IEEE, pp. 1-6, 2019.
[17]T.Bhekisipho, and E. Molloy. “On effectively predicting autism spectrum disorder therapy using an ensemble of classifiers”, Scientific Reports, Vol. 13, no. 1, pp. 19957, 2023. 
[18]Understanding VGG16: Concepts, architecture, and performance. (2023, May 22). Datagen. https://datagen.tech/guides/computer-vision/vgg16/
[19]G. Carneiro, J. C. Nascimento, and A. P. Bradley, “Deep Learning Models for Classifying Mammogram Exams Con-taining Unregistered Multi-View Images and Segmentation Maps of Lesions11”, Deep learning for medical image analysis, pp. 321–339, 2017. 
[20]B. Krishnamurthy, “An introduction to the RELU activation function”, Built In. https://builtin.com/machine-learning/relu-activation-function, 2024.
[21]S. Saxena, Introduction to batch normalization. Analytics Vidhya, 2024, https://www.analyticsvidhya.com/blog/2021/03/introduction-to-batch-normalization/#:~:text=Batch%20normalization%20works%20by%20normalizing,not%20follow%20the%20original%20distribution.
[22]R. Sadik, S. Anwar, and M. L. Reza, “Autismnet: Recognition of autism spectrum disorder from facial expressions using mobilenet architecture”, International Journal, vol. 10, no. 1, pp. 327-334, 2021.
[23]S. J. Pan, and Q. Yang, “A survey on transfer learning”, IEEE Transactions on knowledge and data engineering, vol. 22, no. 10, pp. 1345-1359, 2009.
[24]Autism_Image_Data. (2020, March 21). Kaggle. https://www.kaggle.com/datasets/cihan063/autism-image-data
[25]PyTorch. (n.d.). PyTorch. https://pytorch.org/
[26]Guest_Blog, A Beginner’s Guide to Focal Loss in Object De-tection! Analytics Vidhya, 2020, https://www.analyticsvidhya.com/blog/2020/08/a-beginners-guide-to-focal-loss-in-object-detection/#:~:text=Focal%20loss%20explanation,-Fo-cal%20loss%20is&text=(1%2D%20pt)%CE%B3%20to%20the%20cross%2Dentropy,log%20log(pt).
[27]D. Godoy, “Understanding binary cross-entropy / log loss: a visual explanation”, Medium, 2022, https://towardsdatascience.com/understanding-binary-cross-entropy-log-loss-a-visual-explanation-a3ac6025181a
[28]ImageFolder — Torchvision main documentation. (n.d.). https://pytorch.org/vision/main/generated/torchvision.datasets.ImageFolder.html
[29]GfG. (2024, March 11). ML Underfitting and overfitting. GeeksforGeeks. https://www.geeksforgeeks.org/underfitting-and-overfitting-in-machine-learning/
[30]K. K. M. Rahman, and M. M. Subashini, “Identification of autism in children using static facial features and deep neu-ral networks”, Brain Sciences, vol. 12, no. 1, pp. 94, 2022. https://doi.org/10.3390/brainsci12010094