New Way to Detect Epileptic Seizures
From: Concordia University  - 03/23/2011
By: Sylvain-Jacques Desjardins

Concordia University researchers have developed a computer-based method using
electroencephalograms (EEGs) to detect epileptic seizures as they happen,
which could lead to new insights on the brain's electrical activity. Seizures
have certain recognizable patterns that look like spikes on an EEG recording.
"EEG recordings may cover a period of several weeks," says Concordia
professor Rajeev Agarwal. "That's a lot of data to review. Automating the
process is difficult, because there's no exact definition for a seizure, so
there's no template to look for." The researchers, led by PhD candidate
Rajeev Yadav, developed an algorithm that measures the angles of the spikes
on an EEG recording, with a series of sharp spikes indicating a seizure. In
the study of EEG recordings of seven epileptic patients, the method detected
every seizure with a very low rate of false positives. "Our method may allow
health professionals to gain a much clearer picture of patients’ brain
function," Agarwal says. 

Read the entire article at:
http://www.concordia.ca/now/for-media/press-releases/20110323/new-way-to-detect-epileptic-seizures.php

Links:
Concordia Department of Electrical and Computer Engineering
http://www.ece.concordia.ca/

A novel morphology-based classifier for automatic detection of epileptic
    seizures 
http://ieeexplore.ieee.org/xpl/freeabs_all.jsp?arnumber=5626781

Rajeev Agarwal
http://www.encs.concordia.ca/resources/faculty-and-staff-directory/details/?id=Agarwal%2C%20Rajeev&d=03