Device detects silent seizures by converting EEG data into sound
Stanford researchers have developed an algorithm that translates the brain’s electrical activity into sounds and reliably detect so-called silent seizures, a neurological condition in which patients have epileptic seizures without any of the associated physical convulsions.
About 25 percent of the neurological patients and cardiac arrest patients in intensive care units have non-convulsive seizures, which can lead to permanent brain injury, according to Josef Parvizi, MD, a professor of neurology at the Stanford University Medical Center.
“We are dealing with a very large unmet need—namely, we have lots of patients that are lying quietly and silently in intensive care units around the country seizing,” contends Parvizi. “The evidence is very clear that these patients will suffer in terms of brain health. We don’t have enough EEG machines and enough personnel to go and record the brain waves and interpret them.”
To help address this shortfall, Stanford researchers devised a so-called “brain stethoscope” that detects non-convulsive epileptic seizures that convert brain waves into sound, so that even non-specialists can hear the difference between normal brain activity and a seizure.
“This technology will enable nurses, medical students and physicians themselves to actually assess their patient right there and they will be able to determine if the patient is having silent seizures,” contends Parvizi.
Last week, researchers published results of a proof-of-concept in Epilepsia—a peer-reviewed medical journal focusing on epilepsy—demonstrating that untrained clinicians can differentiate seizures with precision using the technology.
“Our study confirms that individuals without EEG training can detect ongoing seizures or seizure‐like rhythmic periodic patterns by listening to sonified EEG,” conclude Parvizi and his co-authors. “Although sonification of EEG cannot replace the traditional approaches to EEG interpretation, it provides a meaningful triage tool for fast assessment of patients with suspected subclinical seizures.”
According to Parvizi, the device was approved last year by the Food and Drug Administration and is currently being used in several hospitals. It’s called the Ceribell Pocket EEG Device and is designed to record and store EEG signals and to present them in visual and audible formats in real time to help make neurological diagnoses.
Parvizi, who is chairman and co-founder of medical device start-up Ceribell, notes that the algorithm was developed by Chris Chafe, a professor of music at Stanford. “Where do you find a professor of music work with a neurologist to come up with a solution for an unmet clinical need?” he asks.
Going forward, Parvizi says a large clinical trial involving hundreds of patients is currently underway and that results of a smaller clinical trial have been completed and will be published soon.