Optimization of closed-loop electrical stimulation enables robust cerebellar-directed seizure control

Abstract

Additional treatment options for temporal lobe epilepsy are needed, and potential interventions targeting the cerebellum are of interest. Previous animal work has shown strong inhibition of hippocampal seizures through on-demand optogenetic manipulation of the cerebellum. However, decades of work examining electrical stimulation – a more immediately translatable approach – targeting the cerebellum has produced very mixed results. We were therefore interested in exploring the impact that stimulation parameters may have on seizure outcomes. Using a mouse model of temporal lobe epilepsy, we conducted on-demand electrical stimulation of the cerebellar cortex, and varied stimulation charge, frequency, and pulse width, resulting in over a thousand different potential combinations of settings. To explore this parameter space in an efficient, data-driven, manner, we utilized Bayesian optimization with Gaussian process regression, implemented in Matlab with an Expected Improvement Plus acquisition function. We examined three different fitting conditions and two different electrode orientations. Following the optimization process, we conducted additional on-demand experiments to test the effectiveness of selected settings. Regardless of experimental setup, we found that Bayesian optimization allowed identification of effective intervention settings. Additionally, generally similar optimal settings were identified across animals, suggesting that personalized optimization may not always be necessary. While optimal settings were effective, stimulation with settings predicted from the Gaussian process regression to be ineffective failed to provide seizure control. Taken together, our results provide a blueprint for exploration of a large parameter space for seizure control, and illustrate that robust inhibition of seizures can be achieved with electrical stimulation of the cerebellum, but only if the correct stimulation parameters are used.

The little brain and the seahorse: Cerebellar-hippocampal interactions

“…However, results of studies using electrical stimulation of the cerebellum have been very mixed – some studies showed promising effects, but other studies showed no effect of electrical stimulation on seizures, or in some cases even a worsening of the epilepsy phenotype (recently reviewed in Streng and Krook-Magnuson, 2020b ). Reasoning that the differences in outcome may have been in large part due to differences in stimulation parameters, our lab re-examined the potential for electrical stimulation of the cerebellum as a potential strategy to inhibit hippocampal seizures ( Stieve et al, 2022 ). Given the large number of potential combinations to test, we turned to Bayesian optimization, allowing a data-driven approach to test over 1000 stimulation parameter combinations ( Figure 10 ; Stieve et al, 2022 ).…”

LINCs Are Vulnerable to Epileptic Insult and Fail to Provide Seizure Control via On-Demand Activation

“…However, the substantial decrease in the amplitude of LINC-mediated inhibition over the course of the stimulation train may contribute to the poor seizure suppression observed. While it is unclear whether such fatigue could be overcome to allow for more successful intervention strategies, future work could potentially apply methods such as Bayesian optimization to examine whether LINC-mediated seizure suppression could be improved with different stimulation parameters ( Stieve et al, 2023 ). Regardless of why LINC activation was unable to provide meaningful seizure control, and regardless of whether slightly better inhibition could have been achieved through different experimental methods, our results strongly indicate that on-demand activation of LINCs for seizure control is less robust than other approaches already demonstrated to be effective.…”

Optogenetic activation of the superior colliculus attenuates spontaneous seizures in the pilocarpine model of temporal lobe epilepsy

“…It also differs significantly from the effects of focal pharmacological activation, which would provide near‐constant stimulation of the region. Recent work focused on cerebellar stimulation in the intrahippocampal kainate model of temporal lobe epilepsy 32 revealed that Bayesian optimization of stimulation parameters on a subject‐by‐subject basis produced robust and significant improvement in responses to closed‐loop modulation. Further exploration of the parameter space for stimulation within the DLSC can reasonably be expected to do the same.…”

“…We focused initially on open‐loop neuromodulation, as in our prior study in acute models; effects were most robust when stimulation preceded seizure onset. However, closed‐loop neuromodulation is used clinically 40 and has been explored extensively in the preclinical literature 32,41–47 . Future experiments evaluating closed‐loop modulation may reveal a different pattern against chronic, spontaneous seizures.…”

“…However, closed-loop neuromodulation is used clinically 40 and has been explored extensively in the preclinical literature. 32,[41][42][43][44][45][46][47] Future experiments evaluating closedloop modulation may reveal a different pattern against chronic, spontaneous seizures. Second, what effect does chronic neurostimulation have on general behavior, EEG, and sleep?…”