BGU develops wearable advanced warning system for epileptic seizures
The device can generate an advanced warning for the wearer by predicting an oncoming seizure about an hour before it even starts.
By JERUSALEM POST STAFFDr. Oren Shriki, the Department of Cognitive and Brain Sciences at Ben-Gurion University(photo credit: DANI MACHLIS/BEN-GURION UNIVERSITY OF THE NEGEV)
Researchers at the Ben-Gurion University of the Negev (BGU) have developed a wearable device that can predict upcoming epileptic fits for those prone to seizures.The device can generate an advanced warning for the wearer by predicting an oncoming seizure about an hour before it even starts.BGU has licensed its innovative technology to NeuroHelp so that they may advance the technology further, eventually commercializing it. Neurohelp is an Israeli start-up founded by BGN Technologies, incubated within BGU's Oazis accelerator.In the US alone, at least 3.4 million adults and children are living with epilepsy, and rates of new diagnoses are rising, according to the Centers for Disease Control and Prevention.According to BGU's research, around 30% of patients do not properly respond to drugs used to control epilepsy and therefore live not knowing when their next seizure might be. BGU hopes the wearable device will be a viable solution to help improve their quality of life and reduce risk of injury."Epilepsy that is not adequately controlled by medication is prevalent, amounting up to 30% of epilepsy cases, and therefore an accurate, easy to use seizure predicting device is a highly necessary, unmet medical need," said chairperson of NeuroHelp Dr. Hadar Ron."Current seizure alert devices can detect seizures while they are happening, and most of them depend on changes in movement, such as muscle spasms or falls," Ron said. 'Epiness is unique in that it can predict an upcoming seizure and allow the patients and their caretakers to take precautionary actions and prevent injuries."Epiness uses "EEG-based monitoring of brain activity together with proprietary machine-learning algorithms" to predict oncoming seizures without the need for exorbitant amounts of EEG electrodes, by optimizing electrode placement along the scalp.The "machine-learning algorithms are designed to filter noise that is not related to brain activity, extract informative measures of the underlying brain dynamics, and distinguish between brain activity before an expected epileptic seizure and brain activity when a seizure is not expected to occur," BGU further explained.Current seizure prediction devices can alert a seizure in real time, but BGU's solution is the first to give an hour of advanced warning.
"Epileptic seizures expose epilepsy patients to various preventable hazards, including falls, burns and other injuries," said Department of Cognitive and Brain Sciences at BGU and NeuroHelp's scientific founder Dr. Oren Shriki, who led the research at BGU. "Unfortunately, currently there are no seizure-predicting devices that can alert patients and allow them to prepare for upcoming seizures."We are therefore very excited that the machine-learning algorithms that we developed enable accurate prediction of impending seizures up to one hour prior to their occurrence," Shriki added. "Since we have also shown that our algorithms enable a significant reduction in the number of necessary EEG electrodes, the device we are developing is both accurate and user friendly. We are currently developing a prototype that will be assessed in clinical trials later this year."The device's algorithm was tested and developed using a large data set of EGG data from those who suffer from epilepsy."The patient data were divided into short segments that were either preictal [pre-seizure] or inter-ictal. Several machine learning algorithms with differing complexities were trained on pre-allocated training data (comprising 80% of the initial EEG data), and their prediction performance as well as electrode-dependent performance was assessed on the remaining 20% of the data," BGU explained. "The algorithm with the best prediction performance reached a 97% level of accuracy, with near optimal performance maintained (95%) even with relatively few electrodes."
