Abstracts

Rationale: As epilepsy affects approximately one percent of the world population, electrical stimulation of the brain has recently shown potential for additive seizure control therapy. In this study we applied noninvasive transcranial focal stimulation (TFS) via tripolar concentric ring electrodes (TCRE) on the scalp of rats after inducing seizures with pentylenetetrazole (PTZ). We developed a system to detect seizures and automatically trigger the TFS. We evaluated the system on the electrographic activity from rats.Methods: Adult male Sprague-Dawley rats were used in the current study. Three custom TCREs were placed on the scalp with conductive paste and adhered with dental cement. Seizure detection was based on a combination of cumulative sum algorithm (CUSUM) and general likelihood ratio test (GLRT). The real-time CUSUM/GLRT based seizure detection was trained on individual baseline electrographic activity for each rat. All other detector parameters were selected from the training group of rats (n=3). Detection accuracy was calculated for the test group (n=6) of which three rats were used as controls and the other three received TFS (50 mA, 200 s, 300 Hz, 2 m, biphasic, charge-balanced pulses). To validate the seizure detection data were collected for each rat in the following way: first, 5 min of baseline electroencephalogram (EEG) were recorded to train the seizure detector. Next, the seizure detector was activated for 5 min of sham seizure activity (baseline) recording. Finally, seizures were induced with PTZ (55 mg/kg i.p.) and EEG recording continued for another 15 min. Three rats received automatically triggered TFS. Detection accuracy was calculated for periods of sham and real seizure until the first observed myoclonic jerk (MJ) with exception of a 20 s handling period corresponding to the PTZ injection. Sensitivity, specificity and overall accuracy were calculated for time windows of 5 s. The GLRT was also used to compare the average power of electrographic seizure activity for the TFS treated and control rats. Eleven minutes long segments were selected for each rat starting 3 min after the first valid seizure detection accounting for duration of TFS for treated rats.Results: An average seizure onset detection accuracy of 75.2% with sensitivity of 46.6% and specificity of 88.4% was obtained for the test set (n=6). The high specificity is due to few false detections. At the same time seizure onset was detected for all rats prior to the first MJ. The GLRT also showed that TFS significantly (p<0.001) reduced the power of the electrographic seizure activity in the TFS treated group compared to controls in 83.3% of the cases.Conclusions: Automatic detection of electrographic seizure activity was accomplished in all rats well in advance of the behavioral seizure activity. We also confirmed the ability of automatically triggered TFS to significantly reduce the electrographic seizure activity power further suggesting its anticonvulsant effect and the potential of automatic seizure control systems.