Artificial intelligence applied for the diagnosis of absence epilepsy with simultaneously tested patient’s consciousness level in ictal event

Background. Given the difficulties in identifying absences and assessing the level of consciousness in epilepsy patients, it is extremely relevant to develop digital programs for automatic registration and testing of this type of epileptic seizures and related electroencephalographic (EEG) patterns, including those based on artificial intelligence.

Objective: development of an algorithm for automatic detection of absence seizures to test real time patient's consciousness level during long-term video-EEG monitoring.

Material and methods. The work on creating an algorithm was carried out during joint doctor/engineer cooperation. Doctors prepared a set of labeled EEG recordings of patients with verified absence epilepsy. Two independent experts in the generated examinations database mapped typical episodes of absence seizures that allowed to develop training and testing samples for a neural network algorithm to detect EEG absence epiactivity. Next, trained neural network was incorporated into Neuron- Spectrum.NET software to compare its accuracy with similar approaches published elsewhere.

Results. A neural network algorithm was developed and trained using a mapped database to detect EEG absence epiactivity. A comparative analysis of the effectiveness for the proposed method vs. other approaches showed that the former is comparable in quality, whereas in some aspects – even superior to the latter. Accuracy was assessed using a publicly available database with mapped epiactivity episodes.

Conclusion. A hardware and software system for automated assessment of patient’s consciousness level during absence seizure in continuous video-EEG monitoring was proposed. Potentially, neural networks may be applied not only to assess patient’s consciousness level, but also to stop stimulation-mediated seizure onset in the future.

1. Proposal for revised clinical and electroencephalographic classification of epileptic seizures. From the Commission on Classification and Terminology of the International League Against Epilepsy. Epilepsia. 1981; 22 (4): 489–501. http://doi.org/10.1111/j.1528-1157.1981.tb06159.x.

2. Sadleir L.G., Scheffer I.E., Smith S., et al. EEG features of absence seizures in idiopathic generalized epilepsy: impact of syndrome, age, and state. Epilepsia. 2009; 50 (6): 1572–8. http://doi.org/10.1111/j.1528-1167.2008.02001.x.

3. Engel J. Jr. Report of the ILAE сlassification сore group. Epilepsia. 2006; 47 (9): 1558–68. http://doi.org/10.1111/j.1528-1167.2006.00215.x.

4. Hirsch E., French J., Scheffer I.E., et al. ILAE definition of the idiopathic generalized epilepsy syndromes: position statement by the ILAE Task Force on Nosology and Definitions. Epilepsia. 2022; 63 (6): 1475–99. http://doi.org/10.1111/epi.17236.

5. Specchio N., Wirrell E.C., Scheffer I.E., et al. International League Against Epilepsy classification and definition of epilepsy syndromes with onset in childhood: position paper by the ILAE Task Force on Nosology and Definitions. Epilepsia. 2022; 63 (6): 1398–442. http://doi.org/10.1111/epi.17241.

6. Proposal for revised classification of epilepsies and epileptic syndromes. Commission on Classification and Terminology of the International League Against Epilepsy. Epilepsia. 1989; 30 (4): 389–99. http://doi.org/10.1111/j.1528-1157.1989.tb05316.x.

7. Kessler S.K., Shinnar S., Cnaan A., et al. Pretreatment seizure semiology in childhood absence epilepsy. Neurology. 2017; 89 (7): 673–9. http://doi.org/10.1212/WNL.0000000000004226.

8. Hermann B., Jones J., Dabbs K., et al. The frequency, complications and aetiology of ADHD in new onset paediatric epilepsy. Brain. 2007; 130 (Pt. 12): 3135–48. http://doi.org/10.1093/brain/awm227.

9. Fisher R.S., Cross J.H., French J.A., et al. Operational classification of seizure types by the International League Against Epilepsy: position paper of the ILAE Commission for Classification and Terminology. Epilepsia. 2017; 58 (4): 522–30. http://doi.org/10.1111/epi.13670.

10. Gotman J. Automatic detection of seizures and spikes. J Clin Neurophysiol. 1999; 16 (2): 130–40. http://doi.org/10.1097/00004691-199903000-00005.

11. Giannakaki K., Giannakakis G., Vorgia P., et al. Automatic absence seizure detection evaluating matching pursuit features of EEG signals. In: 2019 IEEE 19th International Conference on Bioinformatics and Bioengineering (BIBE). Athens, Greece; 2019: 886–9. http://doi.org/10.1109/BIBE.2019.00165.

12. Glaba P., Latka M., Krause M.J., et al. Absence seizure detection algorithm for portable EEG devices. Front Neurol. 2021; 12: 685814. http://doi.org/10.3389/fneur.2021.685814.

13. Petersen E.B., Duun-Henriksen J., Mazzaretto A., et al. Generic singlechannel detection of absence seizures. In: 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Boston, MA, USA; 2011: 4820–3. http://doi.org/10.1109/IEMBS.2011.6091194.

14. Li L., Zhang H., Liu X., et al. Detection method of absence seizures based on Resnet and bidirectional GRU. Acta Epileptologica. 2023; 5: 7. http://doi.org/10.1186/s42494-022-00117-w.

15. Asif U., Roy S., Tang J., Harrer S. SeizureNet: a deep convolutional neural network for accurate seizure type classification and seizure detection. ArXiv. 2019: abs/1903.03232.

16. Klem G.H., Lüders H.O., Jasper H.H., Elger C. The ten-twenty electrode system of the International Federation. The International Federation of Clinical Neurophysiology. Electroencephalogr Clin Neurophysiol. 1999; 52: 3–6.

17. Seeck M., Koessler L., Bast T., et al. The standardized EEG electrode array of the IFCN. Clin Neurophysiol. 2017; 128 (10): 2070–7. http://doi.org/10.1016/j.clinph.2017.06.254.

18. Shah V., von Weltin E., Lopez S., et al. The Temple University Hospital Seizure Detection Corpus. Front Neuroinform. 2018; 12: 83. http://doi.org/10.3389/fninf.2018.00083.

19. The TUH EEG Seizure Corpus (TUSZ) – v1.5.2. Available at: https://www.kaggle.com/datasets/psyryuvok/the-tuh-eeg-seizure-corpustusz-v152 (accessed 23.11.2023).

20. Kingma D.P., Ba J. Adam: a method for stochastic optimization. arXiv. 2014: 1412.6980. http://doi.org/10.48550/arXiv.1412.6980.

21. Srivastava N., Hinton G., Krizhevsky A., et al. Dropout: a simple way to prevent neural networks from overfitting. J Mach Learn Res. 2014; 15: 1929–58.

22. Ivanov А.А. Overview of current software capabilities for EEG recording and analyzing. Epilepsia i paroksizmalʹnye sostoania / Epilepsy and Paroxysmal Conditions. 2023; 15 (1): 53–69 (in Russ.). http://doi.org/10.17749/2077-8333/epi.par.con.2023.144.