Personalizing stroke treatment using brain computer interfaces


Personalizing stroke treatment
using brain computer interfaces 

After only one hour of training with a brain computer interface (BCI), there were significant changes in people’s brains suggesting that BCIs could be personalized for treating stroke and other brain disorders. That’s according to new research in The Journal of Physiology published today. A BCI allows humans to control devices using only their thoughts. The technology works by measuring the electrical activity of the brain (using a non-invasive technique called an electroencephalography- EEG) and translating it by means of artificial intelligence (AI) into an action, such as communication with a computer or the movement of a prosthetic arm. But are BCIs just reading out neural information or could they also be changing the brain itself. If so, can these changes be used for therapy?

A collaboration between researchers at the Max Planck Institute for Human Cognitive and Brain Sciences, the Technical University of Berlin in Germany and the Public University of Navarra in Spain, investigated the impact of two different BCIs on the human brain.

One task involved participants imagining that they were moving their arm or foot (a so-called motor imagery task) while the other task involved picking out letters from a screen (a visual task). Not only did they find changes in brain areas specific to the type of task being performed (e.g. visual tasks led to changes in visual areas of the brain), but more importantly these changes happened within a very short period (one hour), and not after weeks of physical training. It is well-known that changes can happen when a physical task is being performed, but changes to the brain due to BCI tasks that are purely mental, has been unheard of so far.

Arno Villringer, one of the authors of the study said that “The spatial specificity of BCI effects (e.g. that a visual task impacts the visual areas in the brain) opens up the opportunity of tailoring BCI-based therapeutic approaches individually. In other words, we may be able to personalize treatment for stroke patients according to where in the brain they have damage.”

Klaus-Robert Müller, machine learning and AI expert adds “Tailoring BCI might allow us to use it in the future for rehabilitation in all sorts of brain disorders. Artificial intelligence is key to enabling this technology, as it allows the BCI application without long patient training.”

© Professional Medical Publications. All rights reserved.