June 20, 2018
Researchers Omid Kavehei, PhD, University of Sydney Nano Institute (Australia), and coauthors have used advanced artificial intelligence (AI) and machine learning to develop a generalized method to predict, without surgical implants, when seizures will occur in individuals living with epilepsy.
Their study utilized 3 data sets from Europe and the United States to develop the predictive algorithm with has a sensitivity of up to 81.4% and a false prediction rate as low as 0.06 per hour (Neural Networks. 2018;105:104).
Dr Kavehei and his team outline in their paper a generalized, patient-specific, seizure-prediction method that can alert individuals within 30 minutes of the likelihood of a seizure. Wearable technology would be attached to an affordable device based on the readily available Raspberry Pi technology that could give a patient a 30-minute warning and percentage likelihood of a seizure. An alarm would be triggered between 30 and 5 minutes before seizure onset, giving patients time to find a safe place, reduce stress, or initiate an intervention strategy to prevent or control the seizure.
Dr Kavehei said an advantage of their system is that the system learns as brain patterns change, requiring minimum feature engineering. This allows for faster and more frequent updates of the information, giving patients maximum benefit from the seizure prediction algorithm.
In addition, the system is unlikely to require regulatory approval, and could easily work with existing implanted systems or medical treatments.
The team noted that next steps include applying the neural networks across much larger data sets of seizure information, improving sensitivity. They plan to develop a physical prototype to test the system clinically with partners at the University of Sydney's Westmead medical campus.
—Amanda Del Signore
For more Annals of Long-Term Care articles, visit the homepage
To view the Annals of Long-Term Care print issue, click here