New Studies Show Promise for Restoring Communication Through Brain-to-Computer Interfaces
Exciting breakthroughs in brain-to-computer interface (BCI) technology could change the lives of people who are unable to speak due to severe paralysis, according to two recent studies published in Nature. In both studies, researchers used brain implants to detect and translate brain signals into sentences on a screen, using algorithms. What’s particularly remarkable is that both research teams were able to achieve this with greater speed and accuracy than previous technologies.
The study conducted at Stanford University involved implanting electrodes into the brain of a patient with amyotrophic lateral sclerosis (ALS) in the areas of the brain associated with speech. The BCI was designed to pick up brain activity when the patient attempted to speak, and an algorithm then associated specific brain activity patterns with the sounds of speech. To train the algorithm, the patient practiced vocalizing or silently mouthing sentences over 25 sessions lasting approximately four hours each.
Similarly, researchers at UC San Francisco and UC Berkeley placed a thin sheet containing electrodes onto the brain of a person with paralysis caused by a brainstem stroke. By having the patient attempt to speak, the algorithm learned to recognize the corresponding brain signals and translate them into facial expressions and modulated speech on a digital avatar.
Although the approaches of the two studies differed slightly, the results were similar in terms of accuracy and speed. The Stanford study achieved an error rate of 9.1 percent for a 50-word vocabulary and 23.8 percent for a 125,000-word vocabulary. After about four months, the Stanford algorithm was able to convert brain signals into words at a rate of approximately 68 words per minute. The algorithm developed by UC San Francisco and Berkeley had a median decoding rate of 78 words per minute, with an error rate of 8.2 percent for a 119-word vocabulary and around 25 percent for a 1,024-word vocabulary.
Significant advancements have been made in communication technology, with a new study showing an error rate of only 23 to 25 percent. This is a vast improvement compared to existing methods that are considered “laborious” and can only achieve a rate of 5 to 15 words per minute, compared to the natural speech rate of 150 to 250 words per minute.
Reaching a speed of 60 to 70 words per minute is a major milestone in this field, according to Edward Chang, a co-author of the study. The study involved two different approaches, making it a promising development.
However, it’s important to note that these studies are still in the proof of concept stage and not yet ready for widespread use. One challenge is that the algorithm used in these treatments requires lengthy training sessions. The researchers are hopeful that future advancements will make the training process less intensive.
Additionally, the technology needs to be user-friendly and accessible for individuals to use at home without requiring extensive caregiver training. There are concerns about the invasiveness of brain implants and the potential for electrode degradation. Furthermore, the long and expensive process of rigorous vetting is necessary before the technology can be made available to consumers.
The studies were conducted on patients who still had some movement capabilities, while individuals with certain conditions like late-stage ALS may be completely paralyzed and rely on methods like eye blinking for communication. More research is needed to determine the effectiveness of this technology for individuals in such conditions.
Overall, these advancements show tremendous potential for improving communication for individuals with neurological conditions. Serious consideration is being given to the next steps in implementing this technology safely and on a wide scale.