Skip to main content

Verified by Psychology Today

Artificial Intelligence

Brain-Computer Interface Speeds Neuroscience Research

New AI improves BCI for treatment of brain and neuropsychiatric disorders.

Source: GDJ/Pixabay

The Mayo Clinic and Google Research have published a new study in PLOS Computational Biology that shows how a new artificial intelligence (AI) algorithm for brain-computer interfaces (BCI) enables a greater understanding of the interactions between different brain regions for a more precise treatment of brain diseases and conditions.

“We propose a convergent paradigm to study brain dynamics, focusing on a single brain site to observe the average effect of stimulating each of many other brain sites,” wrote the study authors.

Brain-computer interfaces aim to restore functionality to those impacted by epilepsy, stroke, spinal cord injuries, amyotrophic lateral sclerosis (ALS), cerebral palsy, narcolepsy, Parkinson’s disease, neuromuscular disorders, and more. In mental health, BCIs are being studied as a possible treatment for depression, anxiety, obsessive compulsive disorder (OCD), and other neuropsychiatric disorders.

Brain-Computer interfaces (BCI), also called brain-machine interfaces (BMI), are systems that connect the human brain to external computing devices. The BCI industry is an emerging market that expected to reach USD 3.7 billion in revenue by 2027, growing at a compound annual growth rate of 15.5 percent from 2020-2027 according to Grand View Research.

“Brain networks have been explored electrophysiologically with a variety of techniques, spanning a variety of spatial scales, such as electroencephalography (EEG), magnetoencephalography (MEG), intracranial EEG (iEEG), and microelectrode local field potentials (LFPs),” wrote the scientists. “Efforts to infer connectivity between brain regions may search for correlated signals in response to supervised perturbation by a behavioral task, or in an unsupervised state (“resting” awake or sleeping).”

One of the challenges that neuroscientists face is understanding how brain networks interact with each other. One method is to deliver brief pulses of electrical current in an area of a patient’s brain while monitoring and measuring the resulting voltage in the other regions of the brain.

“In recent years, a sub-field of neuroscience has matured around systematic stimulation and measurement through implanted (iEEG) arrays of brain surface (electrocorticography, ECoG) or deeply-penetrating (stereoelectroencephalography, SEEG) electrodes, typically called “cortico-cortical evoked potentials” (CCEPs) or, for the special case of short pulses separated by several seconds, “single-pulse electrical stimulation” (SPES),” the researchers wrote.

However, these methods often generate complex, difficult-to-measure high-dimensional data, which is a daunting, time-intensive task for humans to identify meaningful patterns, but well-suited for AI machine learning as an assistive analytical tool. So the researchers created a new type of AI algorithm for this purpose.

“Our framework, aiming to better understand brain connectivity, is grounded in a convergent paradigm, examining a set of temporal voltage responses to stimulation, all measured from the same site,” wrote the researchers. “Each response event is labeled by the site of stimulation. Then, a novel algorithm is applied within this framework to identify canonical temporal response motifs, which we call “basis profile curves” (BPCs).”

The researchers named their newly created AI machine learning algorithm as “basis profile curve identification.” According to the scientists, each basis profile curve can be mapped back to the underlying neuroanatomy in a natural manner and with quantified projection strength from each stimulation site.

“Our technique is demonstrated for an array of implanted brain surface electrodes in a human patient,” the researchers reported. “This framework enables straightforward interpretation of single-pulse brain stimulation data and can be applied generically to explore the diverse milieu of interactions that comprise the connectome.”

According to the researchers, their new artificial intelligence algorithm for brain-computer interfaces may help guide the placement of electrodes to help treat network brain diseases in the future.

Copyright © 2021 Cami Rosso All rights reserved.

More from Cami Rosso
More from Psychology Today
More from Cami Rosso
More from Psychology Today