Memory is commonly affected following a traumatic brain injury and may be long-lasting or even permanent. A new study has found that AI-guided electrical brain stimulation in people with moderate-to-severe traumatic brain injury may successfully treat their memory loss.
Traumatic brain injury (TBI) is an injury to the brain, often resulting from a violent blow or jolt to the head or body. It’s commonly caused by motor vehicle accidents, contact sports, firearms, and falls. The severity of TBI depends on the nature of the assault on the brain and the degree of inflammation and injury to neurons and blood vessels it causes.
Immediately after a TBI, a person may suffer from post-traumatic amnesia (PTA), where they’re confused, disoriented, and can’t remember events after the injury. While PTA is transient and can last from minutes to months, moderate-to-severe TBI can have a lasting effect on memory, especially episodic memory, the ability to recall and mentally re-experience specific episodes from one’s past.
In a new study, researchers from the University of Pennsylvania examined how delivering AI-guided electrical brain stimulation to people with moderate-to-severe TBI might improve their memory.
“These patients are often relatively young and physically healthy, but they face decades of impaired memory and executive function,” said Michael Kahana, lead author of the study.
The current study built upon a 2017 study by researchers at the university, which found that electrical stimulation delivered when a machine-learning model predicted that memory was about to fail can improve memory function. The 2017 study used open-loop stimulation, meaning that the AI applied it without regard to the state of the brain.
In 2018, the researchers tested their AI-guided model on 25 epilepsy patients, monitoring their brain activity in real-time and applying stimulation to the left side of the brain only when memory was expected to fail; that is, closed-loop stimulation. They found that memory and learning performance improved by 15%.
For the current study, the researchers recruited eight patients with intractable epilepsy, which is epilepsy uncontrolled by medication, and a history of moderate-to-severe TBI who were undergoing seizure monitoring using implanted intracranial electrodes. Each participant performed a delayed recall test, where they were asked to recall a list of serially presented nouns. The recall task was performed with and without closed-loop brain stimulation.
During the session without stimulation, the researchers identified personalized biomarkers of successful memory encoding that they could use in later sessions to control the closed-loop brain stimulation. During the stimulation session, the machine learning algorithm decoded memory success in real time and triggered stimulation when the participant’s memory performance dipped below their predicted average.
When the researchers examined recall rates with closed-loop stimulation, they found that participants recalled items more reliably than when no stimulation was used. On average, closed-loop stimulation improved memory recall by 19%.
The researchers say their study demonstrates a proof-of-concept for using closed-loop brain stimulation to treat memory loss in patients with TBI.
“By demonstrating therapeutic efficacy in patients who have both a history of moderate-to-severe TBI and documented memory impairment, we hope our findings will accelerate the development of technologies for patients with acquired brain injuries, which could restore some degree of their lost memory function as they attempt to rebuild their post-injury lives,” the researchers said.
The study was published in the journal Brain Stimulation.
Source: University of Pennsylvania