Date Published: July 22, 2019
Publisher: Public Library of Science
Author(s): Timothée Proix, Mehdi Aghagolzadeh, Joseph R. Madsen, Rees Cosgrove, Emad Eskandar, Leigh R. Hochberg, Sydney S. Cash, Wilson Truccolo, Maurice J. Chacron.
The apparent unpredictability of epileptic seizures has a major impact in the quality of life of people with pharmacologically resistant seizures. Here, we present initial results and a proof-of-concept of how focal seizures can be predicted early in advance based on intracortical signals recorded from small neocortical patches away from identified seizure onset areas. We show that machine learning algorithms can discriminate between interictal and preictal periods based on multiunit activity (i.e. thresholded action potential counts) and multi-frequency band local field potentials recorded via 4 X 4 mm2 microelectrode arrays. Microelectrode arrays were implanted in 5 patients undergoing neuromonitoring for resective surgery. Post-implant analysis revealed arrays were outside the seizure onset areas. Preictal periods were defined as the 1-hour period leading to a seizure. A 5-minute gap between the preictal period and the putative seizure onset was enforced to account for potential errors in the determination of actual seizure onset times. We used extreme gradient boosting and long short-term memory networks for prediction. Prediction accuracy based on the area under the receiver operating characteristic curves reached 90% for at least one feature type in each patient. Importantly, successful prediction could be achieved based exclusively on multiunit activity. This result indicates that preictal activity in the recorded neocortical patches involved not only subthreshold postsynaptic potentials, perhaps driven by the distal seizure onset areas, but also neuronal spiking in distal recurrent neocortical networks. Beyond the commonly identified seizure onset areas, our findings point to the engagement of large-scale neuronal networks in the neural dynamics building up toward a seizure. Our initial results obtained on currently available human intracortical microelectrode array recordings warrant new studies on larger datasets, and open new perspectives for seizure prediction and control by emphasizing the contribution of multiscale neural signals in large-scale neuronal networks.
A third of patients with epilepsy are not responsive to medication and experience recurrent seizures through their lives . This number has not changed much in the past decades despite the efforts toward the development of new antiepileptic drugs . The alternative of regional surgical resection to remove identified epileptogenic areas in the brain is only applicable to about 25% of these cases, carry substantial risks, and has limited efficacy . More recently, various studies and clinical trials have started examining new therapeutic approaches based on seizure prediction or early detection for warning systems and seizure prevention and control via closed-loop electrical stimulation [4–6]. These new approaches have relied on intracranial grids covering the cortex over several centimeters and/or depth electrodes. Despite promising initial results [4,5,7], efficacy remains limited.
Based on localized microelectrode array recordings of MUA and/or LFPs, we have shown that human focal seizures can be predicted from neocortical activity away from the identified seizure onset areas. We found that features related to low (< 8 Hz) and high (> 50 Hz) LFP frequency bands as well as multivariate features related to multi-channel MUA were the most predictive.