That would be a few weeks ago, when I first started to look at EEG data (brainwaves) of patients with epilepsy. There has been a recent publication that shows good success at using intelligent software to predict 10 minutes in advance (from the EEG) if someone is going to have an epileptic seizure. So, a few weeks ago, using the same data, I asked the question: “Can I use intelligent software to predict a seizure 30 minutes in advance?” Another way of looking at the question is: “If someone is going to have an epileptic seizure in 30min from now, are there any telltale signals in their brainwaves? So, I ran some experiments,and the answer was “sort of” — I think you can get aout 80% accurate prediction with some patients.
Comments