Adaptive Pre-Ictal Prediction of Epileptic Seizures Using Lightweight SLT-SNN Framework Validated on Local Clinical Data

Abstract

Epileptic seizures affect approximately 1% of the global population, causing significant physical, cognitive, and psychological harm. Existing detection methods identify seizures only after onset, offering limited time for clinical intervention. Furthermore, most models rely on publicly available datasets and operate on resource-heavy architectures that cannot be deployed on wearable devices. This research proposes a novel Adaptive Pre-Ictal SLT-SNN framework that predicts epileptic seizures before onset, adapts to individual patient patterns, operates efficiently on low-power devices, and is validated on locally collected clinical EEG data from Pakistan. The proposed model integrates Slantlet Transform (SLT) for enhanced feature extraction with Spiking Neural Networks (SNN) for biologically inspired, energy-efficient classification. A novel adaptive learning mechanism enables the model to personalize predictions without full retraining. Validated on both standard benchmarks (CHB-MIT, UCI-ESR) and local de-identified clinical data, the framework achieves 97.2% pre-ictal prediction accuracy and maintains 94.8% accuracy on unseen patients, while reducing computational complexity by 62% compared to baseline models. This work demonstrates the feasibility of accessible, personalized, and real-time seizure prediction for resource-limited healthcare environments.

REFERENCES

[1] S. Tripathy, S. Biswas, A. Rath, P. K. Sahoo, A. K. Samantaray, and G. Panda, “Enhanced detection of epileptic seizure using hybrid framework of Slantlet Transform and Spiking Neural Network model,” IEEE Access, vol. 13, pp. 184312–184321, 2025, doi: 10.1109/ACCESS.2025.3625784.

[2] A. H. Shoeb and J. V. Guttag, “Application of machine learning to epileptic seizure detection,” in Proc. 27th Int. Conf. Mach. Learn. (ICML), 2010, pp. 975–982.

[3] R. G. Andrzejak, K. Lehnertz, F. Mormann, C. Rieke, P. David, and C. E. Elger, “Indications of nonlinear deterministic and finite-dimensional structures in time series of brain electrical activity: Dependence on recording region and brain state,” Phys. Rev. E, vol. 64, no. 6, Art. no. 061907, 2001, doi: 10.1103/PhysRevE.64.061907.

[4] J. L. McClelland, B. L. McNaughton, and R. C. O’Reilly, “Why there are complementary learning systems in the hippocampus and neocortex: Insights from the successes and failures of connectionist models of learning and memory,” Psychol. Rev., vol. 102, no. 3, pp. 419–457, 1995, doi: 10.1037/0033-295X.102.3.419.

[5] Y. Chen, X. Li, Y. Han, Z. Wang, Q. Li, Z. Li, X. Li, C. Wei, and X. Wu, “HippoTune: A hippocampal associative loop-inspired fine-tuning method for continual learning,” in Proc. Int. Conf. Learn. Represent. (ICLR), 2026.

[6] Z. Sun, D. Su, X. Li, Q. Wang, S. Zhang, and X. Wu, “Adaptive knowledge generation via reinforcement-guided pattern completion for zero-shot visual question answering,” in Proc. ACM Int. Conf. Multimedia Retr. (ICMR), 2026, pp. 1174–1183, doi: 10.1145/3805622.3810578.

[7] X. Li, “Brain-cognition-inspired large model reasoning in complex dynamic scenes,” National Natural Science Foundation of China (NSFC), General Project No. 62576037, 2026–2029.

[8] P. Rajpurkar et al., “Performance gap in medical AI,” npj Digit. Med., vol. 5, no. 1, Art. no. 89, 2022.

[9] J. Zech et al., “Confounding variables in medical machine learning,” IEEE Trans. Med. Imaging, vol. 37, no. 3, pp. 721–732, 2018.

[10] T. Do et al., “Medical visual question answering: A survey and future directions,” IEEE Trans. Artif. Intell., 2024.

[11] R. M. French, “Catastrophic forgetting in connectionist networks,” Trends Cogn. Sci., vol. 3, no. 4, pp. 128–135, 1999.

[12] V. Gummadavelli, U. Imtiaz, P. A. S. H. Kolla, N. S. Bavirisety, and S. R. Kondapalli, “Discovering and repairing blind spots in LLM-as-a-judge evaluation of multi-turn AI systems,” in Proc. 2026 Int. Conf. Intell. Sustain. AI Syst. (ICOSAAS), Bali, Indonesia, 2026, pp. 1611–1618, doi: 10.1109/ICOSAAS68663.2026.11648543.

[13] “Ghost Signals: Ethical RF adversarial testing of consumer alarm ecosystems,” in Data Science and Security, S. Shukla, H. Sayama, K. Tiwari, J. P. George, and J. V. Kureethara, Eds. Cham, Switzerland: Springer, 2026, Lecture Notes in Networks and Systems, vol. 1945, doi: 10.1007/978-3-032-24075-0_21.

[14] F. Amin, “Adaptive cyber risk management framework for critical infrastructures: Integrating IDS, SOC, and CSIRT for resilient threat mitigation,” in Data Science and Security, S. Shukla, H. Sayama, K. Tiwari, J. P. George, and J. V. Kureethara, Eds. Cham, Switzerland: Springer, 2026, Lecture Notes in Networks and Systems, vol. 1945, doi: 10.1007/978-3-032-24075-0_20.

[15] F. Amin, “Adversarial vulnerability assessment of social data models for defensive cyber operations,” in Data Science and Security, S. Shukla, H. Sayama, K. Tiwari, J. P. George, and J. V. Kureethara, Eds. Cham, Switzerland: Springer, 2026, Lecture Notes in Networks and Systems, vol. 1945, doi: 10.1007/978-3-032-24075-0_16.

[16] P. S. Ponduru, P. P. V. Nandanavanam, and S. K. K. Ponduru, “TRUST-ADAPT: Risk-gated reversible test-time learning with conformal shift evidence and selective memory,” Int. J. Sci. Res. Comput. Sci., Eng. Inf. Technol., vol. 11, no. 2, pp. 4011–4028, 2025.

[17] P. S. Ponduru and P. P. V. Nandanavanam, “PRISM-DM: Drift-aware, imbalance-resilient ensemble learning for adaptive data stream mining,” Shodhshayyam, Int. Sci. Refereed Res. J., vol. 5, no. 2, pp. 494–513, 2022.

[18] P. S. Ponduru, “GATE-FS: Generalization-aware tail-robust feature selection with one-standard-error selector tournaments,” Int. J. Sci. Res. Sci., Eng. Technol., vol. 7, no. 3, pp. 531–550, 2020.

[19] M. I. K. Jabed, M. Imran, A. A. Khan, M. Mehedi, A. Islam, and R. Pervez, “Explainable machine learning framework for early heart disease detection using SMOTE and SHAP,” Vascular and Endovascular Review, vol. 9, no. 1, pp. 316–324, 2026.

[20] M. I. K. Jabed, M. Imran, C. A. Gomes, P. S. Ponduru, S. Mandal, and S. Hassan, “Deep learning and explainable AI framework for predicting lung cancer severity,” J. Comput. Sci. Technol. Stud., vol. 7, no. 12, pp. 573–598, 2025.

[21] M. I. K. Jabed, M. P. Ahmed, F. M. Tofa, M. F. Islam, C. A. Gomes, and R. M. Sirazy, “Federated intrusion detection for Internet of Medical Things networks: Differential privacy, non-IID robustness, and cross-device generalization,” J. Comput. Sci. Technol. Stud., vol. 8, no. 8, pp. 303–315, 2026.

[22] A. Al Kium et al., “Health equity and digital disparities in cancer screening and cardiovascular care across socioeconomic and ethnic groups: A systematic review,” Vascular and Endovascular Review, vol. 6, no. 2, pp. 35–44, 2023.

[23] Q. Wu, E. Fokoue, and R. G. A., “Epileptic seizure recognition data set,” UCI Machine Learning Repository, 2017. [Online]. Available: https://archive.ics.uci.edu/dataset/388/epileptic+seizure+recognition.

[25] C. A. Gomes, M. I. K. Jabed, T. Ahammad, S. Mandal, S. Hassan, M. K. Sejan, P. S. Ponduru, M. M. Bhuiyan, and M. L. Jamali, “Operational and financial evidence for generative AI in United States healthcare administrative workflows,” Frontiers in Computer Science and Artificial Intelligence, vol. 4, no. 4, pp. 138–145, 2025, doi: 10.32996/jcsts.2025.4.4.10.