Quickest seizure onset detection in drug-resistant epilepsy

Epilepsy affects 50 million people worldwide, and 30% remain drug-resistant. This has increased interest in responsive neurostimulation, which is assumed to be most effective when administered right at the seizure onset. We propose a novel framework for seizure onset detection that involves (i) constructing statistics from intracranial multichannel EEG signals (iEEG) to distinguish non-ictal vs. ictal states; (ii) modeling the dynamics of these statistics in each state and the state transitions; (iii) developing an optimal control-based “quickest detection” (QD) strategy to estimate the transition time from non-ictal to ictal states from sequential iEEG measurements. The QD strategy minimizes a cost function of detection delay and false positive probability. The solution is a threshold that non-monotonically decreases over time and avoids responding to rare events that normally trigger false positives. We applied QD to a preliminary dataset of two drug-resistant epileptic patients (87h continuous recordings, 34 and 28 electrodes, respectively, 5 seizures), and achieved 100% sensitivity with low false positive rates (0.16 false positive/h).