The goal is to predict the seizures resulting from Right Temporal Lobe Epilepsy in adults ising time window analysis and applying a threshold to relevant features using windowing technique (overlapping and non-overlapping cases).Performance of the model is evaluated using the prediction time gain indicating how early the alarm is getting triggered. This is compared with the clinical start time of the seizure activity.
The data used for the project has been removed from the repository.
First: Lingeshwaran Kannniappan
Supervisor: Dr. Koushik Maharatna
Secondary Supervisor: Dr. Stephen Beeby