CONVOLUTIONAL NEURAL NETWORKS FOR AUTISM SPECTRUM DISORDER DETECTION USING ELECTROENCEPHALOGRAPHY
DOI:
https://doi.org/10.11113/jm.v48.582Keywords:
Autism Spectrum Disorder, EEG, Machine Learning, Convolutional Neural Network, Deep LearningAbstract
Dealing with children with Autism Spectrum Disorder (ASD) is challenging due to their sensory reactions, leading to behavioral issues, self-injury, and safety concerns. Many individuals with ASD also exhibit atypical sensory processing, increasing anxiety and difficulty in daily life. Existing diagnostic tools like the Autism Diagnostic Observation Schedule (ADOS) are subjective, time-consuming, and heavily dependent on trained professionals. This work is presented as a pilot feasibility study, based on EEG spectrograms, to establish a baseline for future large-scale investigations. The EEG data obtained from King Abdulaziz University Hospital, from 16 participants (12 ASD, 4 neurotypical) were preprocessed to remove noise, segmented into 3.5-second windows, and transformed into time-frequency spectrogram images using the Short-Time Fourier Transform (STFT). These spectrograms were classified using both machine learning (ML) models, including Support Vector Machines (SVM), Decision Trees, and Ensemble Methods, and deep learning (DL) Convolutional Neural Networks (CNNs). While ML models achieved moderate accuracy, with Subspace KNN performing best at 90.27%, CNN architectures significantly outperformed them, Model 4 achieving accuracy of 99.89%, demonstrating stability. Smaller batch sizes (32–64) optimized performance, whereas larger batches (128) degraded accuracy by up to 22%. The results highlight the transformative potential of deep learning in automating ASD diagnosis, offering a rapid, and clinically alternative to traditional methods.
References
H. Hodges, C. Fealko, and N. Soares, “Autism spectrum disorder: definition, epidemiology, causes, and clinical evaluation,” Translational Pediatrics, vol. 9, no. S1, pp. S55–S65, Feb. 2020, doi: 10.21037/tp.2019.09.09.
O. Patil and M. Kaple, “Sensory Processing Differences in Individuals with Autism Spectrum Disorder: A Narrative Review of Underlying Mechanisms and Sensory-Based Interventions,” Cureus, Oct. 2023, doi: 10.7759/cureus.48020.
World Health Organization: WHO, “Autism,” Nov. 15, 2023. https://www.who.int/news-room/fact-sheets/detail/autism-spectrum-disorders
“Data and statistics on autism spectrum disorder,” Autism Spectrum Disorder (ASD), May 16, 2024. https://www.cdc.gov/autism/data-research/index.html
Ova, “Number of children diagnosed with autism in Malaysia increased by 663% from 2013 to 2023,” Ova, Jul. 11, 2024. https://ova.galencentre.org/number-of-children-diagnosed-with-autism-in-malaysia-increased-by-663-from-2013-to-2023/
M. Patil, N. Iftikhar, and L. Ganti, “Neuroimaging Insights into Autism Spectrum Disorder: Structural and Functional Brain,” Health Psychology Research, vol. 12, Sep. 2024, doi: 10.52965/001c.123439.
C. Okoye et al., “Early diagnosis of Autism Spectrum Disorder: A review and analysis of the risks and benefits,” Cureus, Aug. 2023, doi: 10.7759/cureus.43226.
G. S. Dichter, “Functional magnetic resonance imaging of autism spectrum disorders,” Dialogues in Clinical Neuroscience, vol. 14, no. 3, pp. 319–351, Sep. 2012, doi: 10.31887/dcns.2012.14.3/gdichter.
E. Conti et al., “Looking for ‘FNIRS signature’ in autism spectrum: A systematic review starting from preschoolers,” Frontiers in Neuroscience, vol. 16, Mar. 2022, doi: 10.3389/fnins.2022.785993.
P. V. N. N. Ranasinghe and M. S. T. Mapa, “Functional connectivity and cognitive decline: a review of rs-fMRI, EEG, MEG, and graph theory approaches in aging and dementia,” Exploration of Medicine, pp. 797–821, Nov. 2024, doi: 10.37349/emed.2024.00256.
P. V. N. N. Ranasinghe and M. S. T. Mapa, “Functional connectivity and cognitive decline: a review of rs-fMRI, EEG, MEG, and graph theory approaches in aging and dementia,” Exploration of Medicine, pp. 797–821, Nov. 2024, doi: 10.37349/emed.2024.00256.
B. Frauscher et al., “Learn how to interpret and use intracranial EEG findings,” Epileptic Disorders, vol. 26, no. 1, pp. 1–59, Dec. 2023, doi: 10.1002/epd2.20190.
S. M. M. Hasan, M. P. Uddin, M. A. Mamun, M. I. Sharif, A. Ulhaq, and G. Krishnamoorthy, “A Machine Learning Framework for Early-Stage Detection of Autism Spectrum Disorders,” IEEE Access, vol. 11, pp. 15038–15057, Dec. 2022, doi: 10.1109/access.2022.3232490.
T. Akter et al., “Machine Learning-Based Models for Early Stage Detection of autism Spectrum Disorders,” IEEE Access, vol. 7, pp. 166509–166527, Jan. 2019, doi: 10.1109/access.2019.2952609.
J. Carrasco, F. Lisón, and A. Weintraub, “RUSBoost: A suitable species distribution method for imbalanced records of presence and absence. A case study of twenty-five species of Iberian bats,” bioRxiv (Cold Spring Harbor Laboratory), Oct. 2021, doi: 10.1101/2021.10.06.463434.
S. Mostafa, W. Yin, and F.-X. Wu, “Autoencoder based methods for diagnosis of autism spectrum Disorder,” in Lecture notes in computer science, 2020, pp. 39–51. doi: 10.1007/978-3-030-46165-2_4.
X.-A. Bi, Y. Wang, Q. Shu, Q. Sun, and Q. Xu, “Classification of autism spectrum disorder using random support Vector Machine Cluster,” Frontiers in Genetics, vol. 9, Feb. 2018, doi: 10.3389/fgene.2018.00018.
M. Li, Y. Jiang, Y. Zhang, and H. Zhu, “Medical image analysis using deep learning algorithms,” Frontiers in Public Health, vol. 11, Nov. 2023, doi: 10.3389/fpubh.2023.1273253.
S. Jayarathna, Y. Jayawardana, M. Jaime, and S. Thapaliya, “Electroencephalogram (EEG) for delineating objective measure of autism spectrum disorder,” in Advances in bioinformatics and biomedical engineering book series, 2019, pp. 34–65. doi: 10.4018/978-1-5225-7467-5.ch002.
E. Abdulhay, M. Alafeef, H. Hadoush, V. Venkataraman, and N. Arunkumar, “EMD-based analysis of complexity with dissociated EEG amplitude and frequency information: a data-driven robust tool -for Autism diagnosis- compared to multi-scale entropy approach,” Mathematical Biosciences & Engineering, vol. 19, no. 5, pp. 5031–5054, Jan. 2022, doi: 10.3934/mbe.2022235.
L. Yuan and J. Cao, “Patients’ EEG Data Analysis via Spectrogram Image with a Convolution Neural Network,” in Smart innovation, systems and technologies, 2017, pp. 13–21. doi: 10.1007/978-3-319-59421-7_2.
Ö. F. Alçіn, S. Siuly, V. Bajaj, Y. Guo, A. Şengu¨R, and Y. Zhang, “Multi-category EEG signal classification developing time-frequency texture features based Fisher Vector encoding method,” Neurocomputing, vol. 218, pp. 251–258, Sep. 2016, doi: 10.1016/j.neucom.2016.08.050.
S. Y. Sen, O. K. Cura, and A. Akan, “Classification of dementia EEG signals by using Time-Frequency images for deep learning,” 2022 Innovations in Intelligent Systems and Applications Conference (ASYU), pp. 1–6, Oct. 2023, doi: 10.1109/asyu58738.2023.10296777.
S. Raghu, N. Sriraam, Y. Temel, S. V. Rao, and P. L. Kubben, “EEG based multi-class seizure type classification using convolutional neural network and transfer learning,” Neural Networks, vol. 124, pp. 202–212, Jan. 2020, doi: 10.1016/j.neunet.2020.01.017.
Md. N. A. Tawhid, S. Siuly, H. Wang, F. Whittaker, K. Wang, and Y. Zhang, “A spectrogram image based intelligent technique for automatic detection of autism spectrum disorder from EEG,” PLoS ONE, vol. 16, no. 6, p. e0253094, Jun. 2021, doi: 10.1371/journal.pone.0253094.
Downloads
Published
How to Cite
Issue
Section
License
Copyright of articles that appear in Jurnal Mekanikal belongs exclusively to Penerbit Universiti Teknologi Malaysia (Penerbit UTM Press). This copyright covers the rights to reproduce the article, including reprints, electronic reproductions or any other reproductions of similar nature.
