Donate to Science & Enterprise

S&E on Mastodon

S&E on LinkedIn

S&E on Flipboard

Please share Science & Enterprise

Challenge Seeks Methods to More Accurately Detect Epilepsy

3-D brain wiring illustration

3-D brain wiring illustration (NIH)

2 June 2014. Mayo Clinic in Rochester, Minnesota and University of Pennsylvania in Philadelphia are seeking through an open competition new techniques to detect epilepsy seizures sooner and more accurately. The Seizure Detection Challenge has a total prize purse of $8,000 and a deadline for submissions of 19 August 2014.

Epilepsy is a neurological disorder where nerve cell activity in the brain is disturbed, causing seizures with symptoms ranging from blank stares to tingling sensations to loss of consciousness. Some 150,000 people develop epilepsy each year in the U.S., according to the Epilepsy Foundation, with 1 in 26 people developing epilepsy in their lifetimes. Despite advances in medications to control seizures, some 500,000 Americans appear to be resistant to current medications and continue to experience seizures.

A promising avenue for therapy is neurostimulation — mild electrical signals directed to the area of disturbance — if applied early enough to stop a seizure before affecting an individual’s normal activities. Once a seizure builds, it may become too widespread to control with neurostimulation, thus early detection and stimulation are essential. However, current technologies to detect oncoming seizures tend to be highly sensitive, often resulting in false positives.

As a result, the Mayo Clinic/Penn team, led by neurologists Gregory Worrell at Mayo Clinic and Brian Litt at Penn — also a biomedical engineer — are seeking technologies using algorithms that can search large, continuous electroencephalography (EEG) data sets returning low rates of false negatives and false positives. The challenge is hosted by Kaggle, an online community of data scientists in academia and industry.

Competitors in the challenge are provided with two sets of EEG data. One set of data is from four dogs with naturally occurring epilepsy cases, where the animals had electrodes implanted and wore an ambulatory monitoring system. The second set of EEG data is from eight human patients with medication-resistant epilepsy, where readings were taken to identify areas of the brain for surgical removal to prevent future seizures.

The Mayo Clinic/Penn team is seeking solutions that can identify the earliest possible point of EEG changes leading to seizures, with the lowest number of false positives. “Predicting seizures, early and accurately is a first step toward controlling them and improving the quality of life for millions of patients with epilepsy worldwide,” says Worrell in a Mayo Clinic statement.

The competition is sponsored by National Institute of Neurological Disorders and Stroke, part of National Institutes of Health, and the American Epilepsy Society.

Read more:

*      *      *

1 comment to Challenge Seeks Methods to More Accurately Detect Epilepsy