Article published by Johns Hopkins
Two new models could solve a problem that’s long frustrated millions of people with epilepsy and the doctors who treat them: how to find precisely where seizures originate to treat exactly that part of the brain.
By helping surgeons decide if and where to operate, the tools developed by Johns Hopkins University researchers and newly detailed in the journal Brain, could help patients avoid risky and often-ineffective surgeries as well as prolonged hospital stays.
“These are underserved patients,” said Sridevi V. Sarma, associate director of Johns Hopkins Institute of Computational Medicine and head of the Neuromedical Control Systems Lab. “We want surgeries to go well, but we also want to prevent surgeries that may never go well.”
Using equations based on machine learning and calculus to reveal patterns in brain activity, the models identify where seizures begin in the brain. And they do it in just minutes.
Typically patients spend five to 14 days hospitalized with electrodes stuck to their heads, while doctors hope that they’ll have a seizure so that surgeons can map the brain, pinpoint the trouble spot, and plan how to remove it.
“This is a new paradigm,” said Joon-Yi Kang, a neurologist at Johns Hopkins Hospital, who co-authored the studies. “We’re getting more insights into specific brain networks. We’re not waiting around for seizures to happen.”