A new algorithm was created to predict when an epileptic person would have their next seizure as a product of a crowdsourcing competition.
It's not easy to live with epilepsy, a brain disorder characterized by a tendency for recurrent seizures. Life often is very restricted for people suffering from the disorder since there is no reliable way to predict when their next seizure will occur.
This new algorithm is an encouraging discovery for epileptic patients because the researchers claim it predicts the onset of a seizure with an 82 percent success rate, reported the National Institutes of Health (NIH). Before this finding, the best algorithm for predicting seizures was about as accurate as flipping a coin.
The new algorithm was created as part of the American Epilepsy Society Seizure Detection Challenge last August on Kaggle.com, a well-known online platform for data prediction competitions, and co-sponsored by the American Epilepsy Society, NIH's National Institute of Neurological Disorders and Stroke (NINDS), and the Epilepsy Foundation.
The challenge consisted of two contests - detection and prediction of seizures - and more than 500 teams from around the world participated.
The prediction contest was won by Michael Hills, a computer engineer from Australia. Hills enjoys participating in online contests to test his skills on the side in machine learning and digital signal processing, reported NIH.
To develop the winning algorithm, he classified various aspects of localized electrical field potential in the brain.
Researchers believe this new algorithm has the potential to change the lives of the 2.3 million adults who are epileptic.
