Detecting seizure
20 Dec 2011

Nervous system toxicity is the most problematic source of adverse drug reactions throughout drug development1. Assessment of proconvulsant risk is therefore an important aspect of safety pharmacology to reduce drug attrition at an early stage and avoid, at the end, withdrawal from the market.

In this context, NOTOCORD is working on an high-performance automated seizure detector. We are currently in the Research & Development phase, which is primarily focused on online seizure detection (as shown in the picture below), followed by spike train and spike detections.

Example of seizure detection on a rat EEG signal.


Our main objective is to provide a robust analyzer for minimizing the number of false positive, while maintaining a high sensitivity. The module will be first dedicated to rats and non-human primates but we are open to other species. The analyzer will have a user-friendly interface containing only key parameters for ease-of-use, and providing a fast learning curve.

Among the new module’s outputs, duration and location of each seizure and spike train will be extracted from the electroencephalogram, as well as spikes amplitude and duration. As no consensus exists concerning a clear definition of a seizure2, the user could choose to consider spike train and seizure as similar events or separate ones.

Combined with a dedicated seizure marks editor derived from RME10a (Reference mark editor for arrhythmia), you will have a powerful tool to detect spikes and seizures, reject, add, and validate any automated detection for your report.

► If you want to participate to this project, please contact us at:


1- Redfern WS, Ewart L, Hammond TG, Bialecki R, Kinter L, Lindgren S, Pollar CE, Roberts R, Rolf MG, Valentin J-P. Impact and frequency of different toxicities throughout the pharmaceutical life cycle. J Pharmacol Toxicol Methods 2010;62(2):e29.
2- Gotman J. A few thoughts on “What is a seizure?”. Epilepsy Behavior 2011;22:S2-S3.