New AI System Predicts Seizures With Near-Perfect Accuracy [Javascript required]:
For the roughly 50 million people worldwide with epilepsy, the exchange of electrical signals between cells in their brain can sometimes go haywire and cause a seizure—often with little to no warning. Two researchers at the University of Louisiana at Lafayette have developed a new AI-powered model that can predict the occurrence of seizures up to one hour before onset with 99.6 percent accuracy.
"Due to unexpected seizure times, epilepsy has a strong psychological and social effect on patients," explains Hisham Daoud, a researcher who co-developed the new model.
Detecting seizures ahead of time could greatly improve the quality of life for patients with epilepsy and provide them with enough time to take action, he says. Notably, seizures are controllable with medication in up to 70 percent of these patients.
Efficient Epileptic Seizure Prediction Based on Deep Learning$, IEEE Transactions on Biomedical Circuits and Systems (DOI: 10.1109/TBCAS.2019.2929053)
Also at Engadget
(Score: 3, Informative) by darkfeline on Sunday November 17 2019, @10:20PM (1 child)
Your article says:
> A fixed parameter setting applied to all cases predicted 82% of seizures with a false prediction rate of 0.16/h.
That's a bit different than 99.6%.
Join the SDF Public Access UNIX System today!
(Score: 2) by JoeMerchant on Monday November 18 2019, @02:49AM
In 2008 I worked with the researchers further developing those algorithms... the whole thing was done on spit and good intentions, if you read in the 2003 abstract: 5 subjects. What is 5 subjects in statistical significance terms? Usually worse than nothing. The data was ridiculously onerous (HIPAA) and expensive to collect, and even more expensive to obtain independent interpretation of - which is why, after 20 years of dicking around with Lyapunov exponents, they had made so precious little progress.
I'm glad that slapping an AI/machine learning label on it got some fresh funding into the process... maybe the new methods even work a little better than the old ones, but... I believe the real key is in collection and application of "big" data. Size matters, usually more than technique.
🌻🌻 [google.com]