Hybrid Deep Learning Framework for Automated Classification of Pediatric ABR Images in Early Hearing Loss Diagnosis

Authors

  • M. Mukhtar Abusnina Department of Medical Engineering, Faculty of Engineering, Alasmarya Islamic University, Zliten, Libya
  • Saja B. Attallah College of Biomedical Engineering, University of Technology, Baghdad, Iraq
  • Ruslan S. Naseef University of Technology, Baghdad, Iraq.
  • Ali S. Abdulmohsin University of Technology, Baghdad, Iraq.
  • Bilal S. Rasheed College of Biomedical Engineering, University of Technology, Baghdad, Iraq.
  • Maryam H. Jasam College of Biomedical Engineering, University of Technology, Baghdad, Iraq.

Keywords:

Auditory Brainstem Response (ABR), Pediatric Hearing Loss, Deep Learning, CNN-LSTM, ResNet

Abstract

Early identification of auditory impairments in infants is crucial to prevent long-term delays in speech, language, and cognitive development. The Auditory Brainstem Response (ABR) test serves as the gold standard for pediatric auditory screening, yet its' manual interpretation demands expert judgment and time. This study proposes a hybrid deep learning framework that integrates Convolutional Neural Networks (CNNs) with Long Short-Term Memory (LSTM) networks to automatically classify ABR waveform images as normal or abnormal. A comparative ResNet-like CNN model was also developed. The dataset included 187 clinically labeled ABR images from infants aged 1–20 months, expanded to over 1,600 samples through targeted augmentation to balance class distribution. The CNN-LSTM model achieved 99.5% validation accuracy, 100% recall, 99.45% F1-score, and 99.11% specificity, outperforming the ResNet-like model (97.8% accuracy). These results demonstrate the effectiveness of the hybrid CNN-LSTM model in automated ABR interpretation and its clinical potential as a scalable diagnostic tool for early detection of pediatric hearing loss in resource-limited environments.

Downloads

Download data is not yet available.

References

Chen, C., Zhan, L., Pan, X., Wang, Z., Guo, X., Qin, H., Xiong, F., Shi, W., Shi, M., Ji, F., Wang, Q., Yu, N., & Xiao, R. (2021). Automatic recognition of auditory brainstem response characteristic waveform based on bidirectional long short-term memory. Frontiers in Medicine, 7, 613708.‏

Haile, L. M., Kamenov, K., Briant, P. S., Orji, A. U., Steinmetz, J. D., Abdoli, A., ... & Rana, J. (2021). Hearing loss prevalence and years lived with disability, 1990–2019: findings from the Global Burden of Disease Study 2019. The Lancet, 397(10278), 996-1009.‏

Islam, M. N., Sulaiman, N., Bari, B. S., Rashid, M., & Mustafa, M. (2022). A hybrid scheme for AEP based hearing deficiency diagnosis: CWT and convoluted k-nearest neighbour (CKNN) pipeline. Neuroscience Informatics, 2(1), 100037.‏

Jin, F. Q., Huang, O., Robler, S. K., Morton, S., Platt, A., Egger, J. R., Emmett, S. D., & Palmeri, M. L. (2023). A hybrid deep learning approach to identify preventable childhood hearing loss. Ear and Hearing, 44(5), 1262.‏

Liang, S., Xu, J., Liu, H., Liang, R., Guo, Z., Lu, M., Liu, S., Gao, J., Ye, Z., & Yi, H. (2024). Automatic recognition of auditory brainstem response waveforms using a deep learning‐based framework. Otolaryngology–Head and Neck Surgery, 171(4), 1165-1171.‏

Liu, Y., Xiang, L., Li, Q., Li, K., Yang, Y., Wang, T., Qin, Y., Fu, X., Zhao, Y., & Gao, C. (2025). Comparison of deep learning models for objective auditory brainstem response detection: A multicenter validation study. Trends in Hearing, 29, 23312165251347773.‏

Ma, J., Choi, S. J., Kim, S., & Hong, M. (2024). Performance comparison of convolutional neural network-based hearing loss classification model using auditory brainstem response data. Diagnostics, 14(12), 1232.‏

Ma, J., Seo, J. H., Moon, I. J., Park, M. K., Lee, J. B., Kim, H., Ahn, J. H., Jang, J. H., Lee, J. D., Choi, S. J., & Hong, M. (2023). Auditory brainstem response data preprocessing method for the automatic classification of hearing loss patients. Diagnostics, 13(23), 3538.‏

McKearney, R. M., & MacKinnon, R. C. (2019). Objective auditory brainstem response classification using machine learning. International journal of audiology, 58(4), 224-230.‏

McKearney, R. M., Simpson, D. M., & Bell, S. L. (2025). Automated wave labelling of the auditory brainstem response using machine learning. International Journal of Audiology, 64(7), 766-771.‏

Özyurt, F., Majidpour, J., Rashid, T. A., Majidpour, A., & Koç, C. (2023). Multi-transfer learning techniques for detecting auditory brainstem response. Applied Acoustics, 212, 109604.‏

Theunissen, S. C., Rieffe, C., Kouwenberg, M., Soede, W., Briaire, J. J., & Frijns, J. H. (2011). Depression in hearing-impaired children. International journal of pediatric otorhinolaryngology, 75(10), 1313-1317.‏

Wimalarathna, H., Ankmnal-Veeranna, S., Allan, C., Agrawal, S. K., Allen, P., Samarabandu, J., & Ladak, H. M. (2021). Comparison of machine learning models to classify auditory brainstem responses recorded from children with auditory processing disorder. Computer methods and programs in biomedicine, 200, 105942.‏

Yong, M., Panth, N., McMahon, C. M., Thorne, P. R., & Emmett, S. D. (2020). How the world’s children hear: A narrative review of school hearing screening programs globally. OTO open, 4(2), 2473974X20923580.‏

Downloads

Published

2026-02-03

Issue

Vol. 3 (2025)

Section

Track Six: Engineering and Applied Sciences

License

Creative Commons License

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite

Abusnina , M. M., Attallah, S. B., Naseef, R. S., Abdulmohsin, A. S., Rasheed, B. S., & Jasam, M. H. (2026). Hybrid Deep Learning Framework for Automated Classification of Pediatric ABR Images in Early Hearing Loss Diagnosis. The Annual Scientific Conference for Undergraduate and Postgraduate Students, Alasmarya Islamic University, 3, 6. 20-37. https://conf.asmarya.edu.ly/index.php/scupgs/article/view/1001

Most read articles by the same author(s)