MSE Research Project Database

Advanced epileptic seizure warning methods


Project Leader: David Grayden
Staff: David Grayden, Anthony Burkitt, Levin Kuhlmann
Collaborators: Mark Cook (Medicine), Dean Freestone (Medicine)
Sponsors: National Health & Medical Research Council
Primary Contact: David Grayden (grayden@unimelb.edu.au)
Keywords: biosignals; computational neuroscience; electroencephalogram EEG; epilepsy; neuroengineering
Disciplines: Biomedical Engineering,Electrical & Electronic Engineering
Domains: Convergence of engineering and IT with the life sciences
Research Centre: Neuroengineering Research Laboratory

This project will develop epileptic seizure prediction methods, warning patients of the likelihood of an impending seizure, so that precautionary measures can be taken. Seizure prediction will be of great clinical significance as it will improve the lives of 33% of epileptic patients who have drug-resistant epilepsy, by warning of impending seizures and potentially allowing acute therapies to prevent seizures, such as electrical stimulation of the brain or targeted drug delivery.

We have a unique opportunity as a result of the first and only clinical trial of an implantable intracranial EEG monitoring system and seizure predictor developed by NeuroVista Corporation. This trial was completed in Melbourne with CI Cook as a lead investigator. Patients were implanted with the system for periods of up to 2.5 years, providing unprecedented amounts of EEG data per individual. The large amount of data means we can robustly evaluate new and existing seizure prediction methods, and quantify the effects of the diurnal cycle and EEG statistics on seizure prediction. This has not been done before.

The NeuroVista trials showed that seizure prediction is possible and useful, but further research is needed for it to work for everyone.
Our goals are:
1. Evaluate established seizure prediction methods using the unique NeuroVista dataset.
2. Develop new features to better discriminate patient-specific seizure-related time periods.
3. Use the diurnal cycle to improve seizure prediction algorithms.
4. Use iEEG statistics to better quantify long-term epilepsy changes relevant to seizure prediction.

This project will develop robust seizure prediction methods, which will have a resounding impact on the management of drug-refractory epilepsy. Our prediction methods will provide seizure warnings that will greatly reduce the anxiety and stress linked to the uncertainty of when a seizure will occur and greatly enhance the quality of life of patients.