State space modeling of multidien cyclical progression of epilepsy

Krishnakant Saboo, Yurui Cao, Vaclav Kremen, Suguna Pappu, Philippa Karoly, Dean Freestone, Mark Cook, Gregory Worrell, Ravishankar Iyer
Proceedings of the 4th Machine Learning for Health Symposium, PMLR 259:861-885, 2025.

Abstract

The risk of seizures in epilepsy fluctuates in cycles with multiday periodicity. The strength of these patient-specific seizure risk cycles can be modulated by disease processes. There is a lack of computational models of epilepsy that describe the progression and modulation of multiday seizure risk cycles. We developed a state space model (SSM) for epilepsy progression that learns individualized multiday seizure risk cycles from intracranial EEG (iEEG) data. To capture the cyclical nature of seizure risk, our model incorporated cyclical dynamics by using a special rotation matrix structure for the state transition matrix. The model learned patient-specific multiday cycles using a novel expectation-maximization algorithm. We evaluated the model on real-world data from one of the longest continuous iEEG recordings in people with epilepsy. The model forecast iEEG and inferred periods of heightened risk of seizures better than or comparable to baseline models, and provided novel insight into biological factors that modulate seizure risk cycles. To demonstrate the value of the model in developing brain stimulation treatment, the proposed SSM was integrated with reinforcement learning to reduce seizure risk in silico. Our model holds significant potential for addressing clinically important problems.

Cite this Paper


BibTeX
@InProceedings{pmlr-v259-saboo25a, title = {State space modeling of multidien cyclical progression of epilepsy}, author = {Saboo, Krishnakant and Cao, Yurui and Kremen, Vaclav and Pappu, Suguna and Karoly, Philippa and Freestone, Dean and Cook, Mark and Worrell, Gregory and Iyer, Ravishankar}, booktitle = {Proceedings of the 4th Machine Learning for Health Symposium}, pages = {861--885}, year = {2025}, editor = {Hegselmann, Stefan and Zhou, Helen and Healey, Elizabeth and Chang, Trenton and Ellington, Caleb and Mhasawade, Vishwali and Tonekaboni, Sana and Argaw, Peniel and Zhang, Haoran}, volume = {259}, series = {Proceedings of Machine Learning Research}, month = {15--16 Dec}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v259/main/assets/saboo25a/saboo25a.pdf}, url = {https://proceedings.mlr.press/v259/saboo25a.html}, abstract = {The risk of seizures in epilepsy fluctuates in cycles with multiday periodicity. The strength of these patient-specific seizure risk cycles can be modulated by disease processes. There is a lack of computational models of epilepsy that describe the progression and modulation of multiday seizure risk cycles. We developed a state space model (SSM) for epilepsy progression that learns individualized multiday seizure risk cycles from intracranial EEG (iEEG) data. To capture the cyclical nature of seizure risk, our model incorporated cyclical dynamics by using a special rotation matrix structure for the state transition matrix. The model learned patient-specific multiday cycles using a novel expectation-maximization algorithm. We evaluated the model on real-world data from one of the longest continuous iEEG recordings in people with epilepsy. The model forecast iEEG and inferred periods of heightened risk of seizures better than or comparable to baseline models, and provided novel insight into biological factors that modulate seizure risk cycles. To demonstrate the value of the model in developing brain stimulation treatment, the proposed SSM was integrated with reinforcement learning to reduce seizure risk in silico. Our model holds significant potential for addressing clinically important problems.} }
Endnote
%0 Conference Paper %T State space modeling of multidien cyclical progression of epilepsy %A Krishnakant Saboo %A Yurui Cao %A Vaclav Kremen %A Suguna Pappu %A Philippa Karoly %A Dean Freestone %A Mark Cook %A Gregory Worrell %A Ravishankar Iyer %B Proceedings of the 4th Machine Learning for Health Symposium %C Proceedings of Machine Learning Research %D 2025 %E Stefan Hegselmann %E Helen Zhou %E Elizabeth Healey %E Trenton Chang %E Caleb Ellington %E Vishwali Mhasawade %E Sana Tonekaboni %E Peniel Argaw %E Haoran Zhang %F pmlr-v259-saboo25a %I PMLR %P 861--885 %U https://proceedings.mlr.press/v259/saboo25a.html %V 259 %X The risk of seizures in epilepsy fluctuates in cycles with multiday periodicity. The strength of these patient-specific seizure risk cycles can be modulated by disease processes. There is a lack of computational models of epilepsy that describe the progression and modulation of multiday seizure risk cycles. We developed a state space model (SSM) for epilepsy progression that learns individualized multiday seizure risk cycles from intracranial EEG (iEEG) data. To capture the cyclical nature of seizure risk, our model incorporated cyclical dynamics by using a special rotation matrix structure for the state transition matrix. The model learned patient-specific multiday cycles using a novel expectation-maximization algorithm. We evaluated the model on real-world data from one of the longest continuous iEEG recordings in people with epilepsy. The model forecast iEEG and inferred periods of heightened risk of seizures better than or comparable to baseline models, and provided novel insight into biological factors that modulate seizure risk cycles. To demonstrate the value of the model in developing brain stimulation treatment, the proposed SSM was integrated with reinforcement learning to reduce seizure risk in silico. Our model holds significant potential for addressing clinically important problems.
APA
Saboo, K., Cao, Y., Kremen, V., Pappu, S., Karoly, P., Freestone, D., Cook, M., Worrell, G. & Iyer, R.. (2025). State space modeling of multidien cyclical progression of epilepsy. Proceedings of the 4th Machine Learning for Health Symposium, in Proceedings of Machine Learning Research 259:861-885 Available from https://proceedings.mlr.press/v259/saboo25a.html.

Related Material