Epilepsy is a frequent chronic neurologic disorder characterised by recurrent seizures, which affects up to an estimated 70 million people worldwide. Electroencephalography (EEG) is the main tool used clinically to diagnose seizures and epilepsy and is commonly used in rodent disease models of epilepsy to study disease development, understand disease mechanisms and evaluate the effects of anticonvulsant drugs and experimental treatments. Increasingly, the field is moving toward identifying disease-modifying actions of drugs necessitating long-term recordings of EEG in rodents such as mice. However, a key bottleneck is that visual annotation of spontaneous seizures in EEG traces is time-consuming and subject to low inter-observer reproducibility and is complicated by different models and seizure types (duration, morphology). Automated seizure detection is a powerful approach to address this problem which, if sufficiently sensitive and specific, would significantly increase the throughput and reliability of seizure quantification.
We have developed the Epi-AI as a publicly available web server. Epi-AI allowing users to submit a single-channel EEG file, which will be annotated by Epi-AI and returned a list of predicted seizures, including the start time, end time and duration of each, and an image of each predicted seizure. Epi-AI has the potential to be beneficial in research by greatly improving the speed, reliability and reproducibility of seizure analysis in rodent EEG data.
If you want to know more about Epi-AI, please visit https://lisda.ucd.ie/Epi-AI/