Facial Image-Based Autism Detection Using A CNN-‎ExtractedFeature Ensemble with Random Forest Classification

  • Authors

    • V. Krishnamoorthy Professor, Dean, R&D, Department of Electronics and Communication Engineering,‎ St. Michael College of Engineering and Technology, Kalayarkoil, Tamil Nadu, India
    • Cuddapah Anitha Associate Professor, Department of Computer Science and Engineering, ‎School of Computing, Mohan Babu University, Tirupati - 517102, Andhra Pradesh, India
    • Swathy V Assistant Professor, Department of Electronics and Communication Engineering, King's Engineering College
    • M. Indirani Assistant Professor, Department of Computer Science and Business Systems, ‎Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College, Avadi, Chennai
    • G. Sasi Professor, Department of Electronics and Communication Engineering,‎ Chettinad Institute of Technology, Chettinad Academy of Research and Education, Manamai off Campus, ‎ECR,‎ Chengalpattu- 603 102, Tamil Nadu, India
    • K. Selvakumarasamy Professor, Department of Electronics and Communication Engineering, ‎Saveetha School of Engineering, Saveetha Institute of Medi-cal and Technical Sciences, ‎SIMATS 602 105 Chennai, India
    https://doi.org/10.14419/4mmwca71

    Received date: July 4, 2025

    Accepted date: August 19, 2025

    Published date: August 31, 2025

  • ASD; CNN; RF; F1-Score; AUC

  • Abstract

    For effective care and attention, autism spectrum disorder (ASD) must be identified early and ‎This study suggests a hybrid model that blends convolutional neural networks (CNN) and ‎random forest (RF) classifiers. The proposed model had a high classification accuracy of ‎‎91.3%, demonstrating its ability to reliably detect cases of ASD. With a precision of 90.7%, ‎the model effectively reduces false positives, and its 89.5% recall ensures excellent sensitivity ‎to actual ASD patients. The balanced F1-Score of 90.1% shows that the model performs ‎consistently in terms of both precision and recall. Additionally, the model's Area Under the ‎Curve (AUC) score of 0.94 indicates its exceptional ability to discriminate between ASD and ‎non-ASD classes. All of these results demonstrate the Hybrid CNN + RF model's strength ‎and dependability, which makes it a promising non-invasive technique for early facial feature-based ASD screening‎.

  • References

    1. Ramya, R., & Arokiaraj, S. (2024). Enhancing autism severity prediction: A fusion of convolutional neural networks and random forest model. In-ternational Journal of Information Technology and Security, 16(2), 111–120. https://doi.org/10.59035/VNWF2548.
    2. Ledhari, M., Rahouti, M., & Alfatemi, A. (2024, November). Towards equitable ASD diagnostics: A comparative study of machine and deep learn-ing models using behavioural and facial data. arXiv preprint arXiv:2411.05880. https://arxiv.org/abs/2411.05880.
    3. Khanna, A., Mishra, M., & Pati, U. C. (2023). A hybrid feature-based approach of facial images for the detection of autism spectrum disorder. In Advances in Computational Intelligence (pp. 413–425). Springer. https://doi.org/10.1007/978-981-99-3878-0_33.
    4. Rahman, K. K. M., & Subashini, M. M. (2022). Identification of autism in children using static facial features and deep neural networks. Brain Sci-ences, 12(1), 1–15. https://doi.org/10.3390/brainsci12010094.
    5. Deepanjali, P., Sreeja, J., & Bhavani, K. (2025). Identification of autism in children using static facial features and deep neural networks. Interna-tional Journal of Engineering Science and Advanced Technology, 25(1), 28–35. https://doi.org/10.36893/IJESAT.2025.V25I1.05.
    6. Alam, M. B., Sultana, M. S., & Shihab, T. I. (2023, September). An autism detection architecture with fusion of feature extraction and classifica-tion. In Proceedings of the International Conference on ICT for Sustainable Development (ICICT4SD) (pp. 1–5). IEEE. https://doi.org/10.1109/ICICT4SD59951.2023.10303542.
    7. Awaji, B., Senan, E. M., & Olayah, F. (2023). Hybrid techniques of facial feature image analysis for early detection of autism spectrum disorder based on combined CNN features. Diagnostics, 13(18), 1–16. https://doi.org/10.3390/diagnostics13182948.
    8. Saranya, A., & Anandan, R. (2022). Facial action coding and hybrid deep learning architectures for autism detection. Intelligent Automation & Soft Computing, 32(1), 1–14. https://doi.org/10.32604/iasc.2022.023445.
    9. Anjum, J., Hia, N. A., & Waziha, A. (2024, May). Deep learning-based feature extraction from children's facial images for autism spectrum disorder detection. In Proceedings of the ACM International Conference (pp. 118–123). ACM. https://doi.org/10.1145/3660853.3660888.
    10. Chern, L. H., & Al-Hababi, A. Y. S. (2024). A novel hybrid UNet-RBF and CNN-RBF algorithm for autism spectrum disorder classification. Journal of Cognitive Sciences and Human Development, 10(1), 25–36. https://doi.org/10.33736/jcshd.6778.2024.
    11. Elshoky, B., Younis, E. M. G., & Ali, A. A. (2022). Comparing automated and non-automated machine learning for autism spectrum disorders clas-sification using facial images. ETRI Journal, 44(4), 526–536. https://doi.org/10.4218/etrij.2021-0097.
    12. Varsha, P., & U. V. (2024, January). Identification of autism spectrum disorder in children from facial features using deep learning. In Proceedings of the International Conference on Advanced Computing Technologies (ICAECT). IEEE. https://doi.org/10.1109/ICAECT60202.2024.10469379.
    13. Mahamood, M. N., Uddin, M. Z., & Shahriar, M. A. (2023, October). Autism spectrum disorder classification via local and global feature represen-tation of facial image. In Proceedings of the IEEE Systems, Man, and Cybernetics (SMC) Conference. IEEE. https://doi.org/10.1109/SMC53992.2023.10394092.
    14. Arumugam, S. R., Balakrishna, R., & Khilar, R. (2021, October). Prediction of autism spectrum disorder in children using face recognition. In Pro-ceedings of the IEEE International Conference on Intelligent Computing and Optimization for Smart Energy and Cognitive Systems (ICOSEC) (pp. 1–5). IEEE. https://doi.org/10.1109/ICOSEC51865.2021.9591679.
    15. Qiu, L., & Zhai, J. (2024). A hybrid CNN-SVM model for enhanced autism diagnosis. PLOS ONE, 19(5), 1–18. https://doi.org/10.1371/journal.pone.0302236.
    16. Liu, R., & He, S. (2021, September). Proposing a system-level machine learning hybrid architecture and approach for a comprehensive autism spec-trum disorder diagnosis. arXiv preprint arXiv:2109.12345. https://arxiv.org/abs/2109.12345.
    17. Khan, B., Akram, A., & Bhatti, S. M. (2024). Autism spectrum disorder detection in children via deep learning models based on facial images. Bul-letin of Business and Economics, 9(2), 90–98. https://doi.org/10.61506/01.00241.
    18. Sakib, M., Atul, S. H. M., & Islam, M. (2024, September). Comparative analysis of ML and DL models for discriminating autistic vs non-autistic spectrum disorder using facial images. In Proceedings of the IEEE COMPAS Conference. IEEE. https://doi.org/10.1109/COMPAS60761.2024.10796221.
    19. Ikermane, M., & Mouatasim, A. E. (2023). Web-based autism screening using facial images and a convolutional neural network. Indonesian Journal of Electrical Engineering and Computer Science, 29(2), 1250–1258. https://doi.org/10.11591/ijeecs.v29.i2.pp1140-1147.
    20. Wadhera, T., Mahmud, M., & Brown, D. J. (2023). A deep concatenated convolutional neural network-based method to classify autism. In Artifi-cial Intelligence Applications (pp. 411–423). Springer. https://doi.org/10.1007/978-981-99-1648-1_37.
    21. Alsaade, F. W., & Alzahrani, M. S. (2022). Classification and detection of autism spectrum disorder based on deep learning algorithms. Computa-tional Intelligence and Neuroscience, 2022, 1–11. https://doi.org/10.1155/2022/8709145.
    22. Parvej, B., Alam, S., & Fahim, F. I. (2024, September). Computer vision-based interactive autism detection system using deep learning. In Proceed-ings of the IEEE COMPAS Conference. IEEE. https://doi.org/10.1109/COMPAS60761.2024.10796046.
    23. Bhise, A., Gawade, S., & Pawar, A. (2024, March). Facial-based autism spectrum disorder diagnosis using enhanced deep learning models. In Pro-ceedings of the IEEE INC Conference. IEEE. https://doi.org/10.1109/InC460750.2024.10649230.
    24. Bhat, C., Goutham, G., & Bannur, C. (2024, April). Deep dive into predictive modeling for autism spectrum disorder using facial features and ma-chine learning. In Proceedings of the IEEE IC3IOT Conference. IEEE. https://doi.org/10.1109/IC3IoT60841.2024.10550416.
    25. Paneru, B., Paneru, B., & Shah, K. B. (2024). Autism spectrum disorder prediction using hybrid deep learning model and a recommendation system application for autistic patient. Journal of Computer Science, 20(9), 1040–1050. https://doi.org/10.3844/jcssp.2024.1040.1050.
    26. Kumar, N. A., & Kamatchi, S. (2025). Predicting heart failure readmission with improved accuracy from discharge summaries by using logistic re-gression and support vector machine algorithm. AIP Conference Proceedings, 3300(1). AIP Publishing. https://doi.org/10.1063/5.0277190.
    27. Sathish, K., Sharma, R. R., Tiwari, M., Sungheetha, A., Pradeep, G. G. S., & Ellappan, V. (2025, May). MultiScale wavelet-based compression schemes for preserving diagnostic information in medical imaging. In International Conference on Sustainability Innovation in Computing and Engi-neering (ICSICE 2024) (pp. 1332–1342). Atlantis Press. https://doi.org/10.2991/978-94-6463-718-2_111.
    28. Sundararajan, S. C. M., et al. (2023). Improved smart healthcare system of cloud-based IoT framework for the prediction of heart disease. Infor-mation Technology and Control, 52(2), 529–540. https://doi.org/10.5755/j01.itc.52.2.32777.
    29. Ramkumar, S., Kumar, M. R., & Sasi, G. (2022). Programmed for automatic bone disorder clustering based on cumulative calcium prediction for feature extraction. Clinical Laboratory, 68, 1579–1587. https://doi.org/10.7754/Clin.Lab.2021.210844.

  • Downloads

  • How to Cite

    Krishnamoorthy , V. ., Anitha , C. ., V, S. ., Indirani , M. ., Sasi , G. ., & Selvakumarasamy , K. . (2025). Facial Image-Based Autism Detection Using A CNN-‎ExtractedFeature Ensemble with Random Forest Classification. International Journal of Basic and Applied Sciences, 14(4), 826-831. https://doi.org/10.14419/4mmwca71