Computer-Aided Diagnosis System for Autism in Children Using Facial Features
Computer-Aided Diagnosis System for Autism in Children
DOI:
https://doi.org/10.63841/iue31611Keywords:
Autism Spectrum Disorder (ASD), Machine Learning, Feature Extraction Techniques, Canny Edge Detection, Discrete Fourier Transform (DFT), Image-Based Classification.Abstract
Autism Spectrum Disorder (ASD) is a complex neurodevelopmental condition characterized by challenges in communication, social interaction, and behavioral patterns. Early diagnosis plays a crucial role in improving developmental outcomes; however, existing diagnostic procedures remain time-intensive, rely heavily on subjective clinical observations, and require extensive professional expertise. This study presents a fully automated computer-aided diagnosis framework that leverages facial-image analysis to support early ASD screening. The proposed approach introduces a hybrid feature-extraction pipeline that integrates Canny edge detection with frequency-domain analysis using the Discrete Fourier Transform (DFT). This combination enhances discriminative facial patterns by preserving salient structural boundaries while capturing global frequency characteristics, an approach that, to the best of our knowledge, has not been previously applied to ASD detection.
A publicly available dataset of pediatric facial images was used, covering both autistic and non-autistic subjects. The images underwent standardized preprocessing, followed by extraction of spatial features (GLCM, LBP) and hybrid edge-enhanced frequency features. Three machine-learning classifiers: Support Vector Machine (SVM), Random Forest (RF), and K-Nearest Neighbor (KNN), were trained and evaluated using multiple performance metrics, including Accuracy, Sensitivity, Specificity. Experimental results demonstrated that the hybrid Canny+DFT feature set significantly improved classification performance, achieving a highest accuracy of 92.5% using the SVM classifier.
Overall, the findings highlight the potential of integrating spatial and frequency information for automated ASD screening using facial imagery. The proposed approach offers a computationally efficient, interpretable, and non- invasive tool that may complement traditional diagnostic assessments and contribute to accessible early intervention strategies.
Downloads
References
A. Klin, W. Jones, R. Schultz, F. Volkmar, and D. Cohen, “Visual fixation patterns during viewing of naturalistic social situations as predictors of social competence in individuals with autism,” Arch. Gen. Psychiatry, vol. 59, no. 9, pp. 809–816, 2002.
S. Baron-Cohen, H. Ring, and E. Tager-Flusberg, “What is autism?,” Lancet, vol. 368, no. 9531, pp. 1445–1446, 2006.
N. Duda, L. Zhang, and C. Li, “Use of machine learning for behavioral diagnosis of autism with limited dataset,” Sci.
Rep., vol. 10, pp. 1–12, 2020.
American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders (DSM-5), 5th ed. Washington, D.C., USA: APA, 2013.
A. Hashemi, L. F. Carpenter, and S. F. Shultz, “Craniofacial morphology in children with autism,” Mol. Autism, vol.
, pp. 1–12, 2015.
H. J. Hammel and F. M. Hildebrandt, “Facial asymmetry in autism spectrum disorder,” J. Autism Dev. Disord., vol.
, pp. 350–363, 2019.
U. R. Acharya, E. Y. Kuan, P. Subbanna, and K. Tamura, “Automated diagnosis of autism using image-based texture
analysis,” Biomed. Signal Process. Control, vol. 68, p. 102689, 2021.
M. R. Rosenfeld, “Facial morphology and neurodevelopmental disorders,” Curr. Opin. Pediatr., vol. 28, no. 6, pp.
–722, 2016.
R. M. Haralick, K. Shanmugam, and I. Dinstein, “Textural features for image classification,” IEEE Trans. Syst., Man,
Cybern., vol. SMC-3, no. 6, pp. 610–621, 1973.
T. Ojala, M. Pietikäinen, and D. Harwood, “A comparative study of texture measures,” Pattern Recogn., vol. 29, no.
, pp. 51–59, 1996.
A. Bovik, Handbook of Image and Video Processing, 2nd ed. New York, USA: Academic Press, 2005.
Kaggle, “Autism facial image dataset,” Available: https://www.kaggle.com/ (accessed: August. 2025).
D. S. Kunda and T. Goel, “Machine learning pipeline for autism detection using facial images,” Pattern Recogn.
Lett., vol. 150, pp. 176–182, 2021.
Y. Guo, Z. Li, and H. Zhou, “Deep multimodal fusion for autism detection from facial cues,” Neurocomputing, vol.
, pp. 257–270, 2022.
H. Al-Shammari, M. Saleh, and R. Al-Mutairi, “Hybrid handcrafted and CNN features for autism detection,” Appl.
Sci., vol. 12, no. 11, p. 5484, 2022.
S. Adriano and M. Broome, “Deep learning for autism diagnosis using facial behavior,” IEEE Trans. Affect.
Comput., early access, 2023.
A. Patel and F. Hossain, “Explainable AI for autism screening from facial images,” Expert Syst. Appl., vol. 233, p.
, 2023.
R. C. Gonzales, R. E. Woods, and S. L. Eddins, Digital Image Processing Using MATLAB. Upper Saddle River, NJ, USA: Pearson Prentice Hall, 2004.
J. Canny, “A computational approach to edge detection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-8, no. 6, pp. 679–698, 1986.
C. Cortes and V. Vapnik, “Support-vector networks,” Machine Learning, vol. 20, no. 3, pp. 273–297, 1995.
Z. Wang, A. Bovik, H. Sheikh, and E. Simoncelli, “Image quality assessment: From error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, 2004.
L. Breiman, “Random forests,” Mach. Learn., vol. 45, pp. 5–32, 2001.
N. Altman, “An introduction to kernel and nearest-neighbor nonparametric regression,” Am. Stat., vol. 46, no. 3, pp.
–185, 1992.
T. W. Loong, “Understanding sensitivity and specificity with the right side of the brain,” British Medical Journal, vol. 327, no. 7417, p. 716, 2003.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Academic Journal of International University of Erbil
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
