By Yosef Scher, Science and Technology Editor
Communication is challenging for people with tetraplegia, a condition that causes individuals to be paralyzed from the neck down. Even paralyzed individuals that can communicate can only do so with minimal movements, such as eye gestures or attempts at mouthing words. However, with the new implant technology developed by a team of researchers at the California Institute of Technology, communication has become accessible for the tetraplegia community.
While a significant amount of research has been done regarding paralyzed people moving objects, until recently, there has been little in the realm of communication and speech. Using a brain-machine interface, Sarah Wandelt, a lead researcher on this project, enabled paralyzed people to reconstruct speech by having them think of what they wanted to say. Brain-machine interfaces “acquire brain signals, analyze them, and translate them into commands that are relayed to output devices that carry out desired actions” and work in tandem with electrodes implanted in a person’s head to help them perform an action. When Wandelt had paralyzed people try to speak, “neural signals associated with words [were] detected by electrodes implanted in the brain. The signals [were] then translated into text, which can be made audible by computer programs that generate speech.”
In order for the brain-machine interface to work properly, it needed to be trained to recognize specific “brain patterns produced when certain words were spoken internally” by the tetraplegic participant. While some people would assume that training a brain-machine interface to recognize internal speech patterns would take a significant amount of time, the device that Wandelt created only needed to be trained for fifteen minutes. After training the brain-machine interface, participants sat in a chair with a screen in front of them. Next, a word appeared on the screen, and the participants were instructed to say the word internally, i.e., think about the word on the screen. Astonishingly, Wandelt’s brain-machine interface algorithms were “able to predict words with an accuracy up to 91 percent,” far surpassing any current brain-interface machines that provide people with tetraplegia with similar abilities.
Wandelt hopes that after enhancing her device, she will be able to extend this new form of communication to patients who do not have tetraplegia but suffer from other diseases, including brain injuries and amyotrophic lateral sclerosis (ALS). Leigh Hochberg, a leading scholar in the usage of neurotechnology to help people with paralysis, felt that Wandelt’s findings provided significant results that could lead to an advancement in effectively helping people with tetraplegia gain their ability to speak. After hearing her speak at the Society of Neuroscience annual meeting, Hochberg commented that Wandelt’s approach is “really exciting, and reinforces the power of bringing together fundamental neuroscience, neuroengineering and machine learning approaches for the restoration of communication and mobility.”
Although this new device is revolutionary in its field, Wandelt feels that many improvements still need to be made, including making it faster and more accurate. Additionally, this device needs further research and development for people with more severe speech disorders. That being said, Wandelt and Hochberg believe that it is still “early days for this technology,” and they are hopeful that with a little more time, people with tetraplegia will not have to worry about not having the ability to speak clearly and effectively.