Melbourne NeuroVista Seizure Prediction Trial

This data is licensed under a Creative Commons license with conditions on ATTRIBUTION, NON-COMMERCIAL use and SHARE-ALIKE

- data use is for private individual/research/teaching training and no other purpose
- commercial entities may not use this data for any reason
- this data may not be used for patent applications, licensed software, IP or other for-profit use


Data should be cited as the Melbourne University NeuroVista Seizure Prediction Data (

The following reference should be used:
Cook, Mark J., et al. "Prediction of seizure likelihood with a long-term, implanted seizure advisory system in patients with drug-resistant epilepsy: a first-in-man study." The Lancet Neurology 12.6 (2013): 563-571.

- derivative works must be distributed under a Creative Commons license
- derivative works may include (but not limited to) neural networks, classifiers, mathematical models, computational models, networks, figures that are derived substantially from the data

Seizures for 12 patients recorded during the NeuroVista trial. Folders contain a number of .mat files (one file per seizure). Each file contains a variable 'data' with dimensions TxN where T is the data length and N is the number of electrode channels (16). T is variable (depending on seizure length. Seizure onset is at 1 minute. Seizure offset is 10s before the end of data. Sampling rate is 400Hz.

For more information refer to Cook et al (2013).
To use this data in your work please cite the original paper (provided in references) by Cook et al (2013).