Authors : PT Chitenhe; E M Hukuimwe; M Gondo

Volume/Issue : Volume 10 - 2025, Issue 7 - July

Google Scholar : https://tinyurl.com/4tpwwxw5

DOI : https://doi.org/10.38124/ijisrt/25jul589

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Note : Google Scholar may take 30 to 40 days to display the article.

Abstract : Epilepsy is a critical neurological disorder, especially prevalent in developing nations like Zimbabwe, where healthcare infrastructure and diagnostic resources remain limited. Electroencephalograms (EEGs) provide essential clinical insight into seizure activity but require expert interpretation, often leading to delays in diagnosis. This research presents a comparative study of three machine learning models—Random Forest (RF), K-Nearest Neighbours (KNN), and Convolutional Neural Networks (CNN)—for the automated detection of epileptic seizures using a Zimbabwe-specific EEG dataset. The dataset comprises 2,216 EEG segments, each with 667 extracted features. The study investigates each model's effectiveness using metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. Among all models, RF achieved the highest classification performance with an accuracy of 85.78%, suggesting strong potential for integration into clinical decision-support tools to aid early and reliable epilepsy diagnosis in under-resourced settings.

Keywords : Epilepsy, EEG, Machine Learning, Random Forest, Convolutional Neural Networks, Seizure Detection, Zimbabwe.

References :

1. Acharya, U. R., et al. (2018). [6] A deep convolutional neural network for automated diagnosis of epilepsy using EEG signals. [6][7] Future Generation Computer Systems, 86, 415-422.
2. Duan, R., et al. (2019). Epileptic seizure detection using CNN-based spectral-spatial feature representation of EEG signals. IEEE Access, 7, 77969-77977.
3. Goyal, M., et al. (2020). Application of machine learning in epileptic seizure detection. Journal of Healthcare Engineering, 2020, 8886297.
4. Khan, Y. U., et al. (2021). Automated seizure detection in EEG signals using ensemble learning. Biomedical Signal Processing and Control, 69, 102922.
5. Li, X., & Chen, W. (2021). Epileptic seizure detection based on EEG signals and CNN. [7][8] Frontiers in Neuroinformatics, 12, 95.
6. Truong, N. D., et al. (2021). A deep learning framework for epileptic seizure detection using EEG signals. [7][9] IEEE Transactions on Neural Systems and Rehabilitation Engineering, 29, 687-696.
7. Ullah, I., et al. (2018). Automated epileptic seizure detection using উন্নত features of EEG signals. Biocybernetics and Biomedical Engineering, 38(4), 734-744.
8. Duta, I. (2025). [5] A cautiously optimistic future for affordable epilepsy diagnosis. Oxford Martin School. https://www.oxfordmartin.ox.ac.uk/blog/a-cautiously-optimistic-future-for-affordable-epilepsy-diagnosis
9. Ifakara Health Institute. [10] (2025). [3][5] Leveraging AI in epilepsy diagnosis. https://www.ihi.or.tz/blog/news/digital-health-leveraging-ai-in-epilepsy-diagnosis/
10. Kabir, M.S.A., Roy, T., Rahimi, M.J. (2022). Robust Classifier Selection for Cost-Effective EEG Seizure Detection in Remote Areas. 2022 2nd International Conference on Computing and Communication Systems (I3CS). IEEE.
11. The IC² Institute. (n.d.). AI-Based Seizure Diagnosis for Low-Resource Areas. https://ic2.utexas.edu/research/ai-based-seizure-diagnosis-for-low-resource-areas/
12. Federated Machine Learning for Epileptic Seizure Detection using EEG. (2024). [11] International Journal of Advanced Computer Science and Applications, 15(4). [11]
13. Jia, K., et al. (2022). Epileptic seizure detection with deep EEG features by convolutional neural network and shallow classifiers. [6] Frontiers in Neuroscience, 16, 840141.
14. Nahzat, F., & Yağanoğlu, M. (2024). [12] Epileptic seizure detection using machine learning on electroencephalogram data. [7][13] Journal of Emerging Investigators, 24(028).
15. Roy, S., et al. (2019). Detection Analysis of Epileptic EEG Using a Novel Random Forest Model Combined With Grid Search Optimization. [14][15] Frontiers in Human Neuroscience, 13, 52.
16. Global, regional, and national burden of epilepsy, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016 - PMC - PubMed Central.
17. Evaluating the generalisability of region-naïve machine learning algorithms for the identification of epilepsy in low-resource settings - PubMed Central.