Sign In

Communications of the ACM

ACM TechNews

Crowdsourcing Algorithms to Predict Epileptic Seizures

View as: Print Mobile App Share:
Different algorithms performed best for different patients.

A study by University of Melbourne researchers reveals clinically relevant epileptic seizure prediction is possible in a wider range of patients than previously thought, thanks to the crowdsourcing of more than 10,000 algorithms worldwide.

Credit: Pixabay

More than 10,000 crowdsourced algorithms can enable clinically relevant epileptic seizure prediction across a broad spectrum of patients, according to a study from the University of Melbourne in Australia.

The algorithms were developed via a contest hosted on the data science competition platform, and based on long-term electrical brain activity recordings acquired in 2013 from the first clinical trial of the implantable NeuroVista Seizure Advisory System.

Melbourne's Levin Kulhmann says the competition attracted more than 646 participants and 478 teams worldwide.

The algorithms can differentiate 10-minute inter-seizure from pre-seizure data clips, with the leading programs tested on patients with the lowest seizure prediction performance based on previous studies.

Kuhlmann says the assessments "revealed on average a 90% improvement in seizure prediction performance, compared to previous results."

From The Melbourne Newsroom
View Full Article


Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA


No entries found