Brain sensor helps decode speech, movement in ALS patient: Study
Brain-computer interfaces may restore communication, independence

Implanting a sensor in a region of the brain that helps to control speech was shown in a U.S. study to not only decode the intended speech of a man with amyotrophic lateral sclerosis (ALS), but also to help this patient control a computer mouse.
According to the researchers, the use of new technology known as brain-computer interfaces, or BCIs, may — once further developed — help to restore communication and independence among people with ALS and other diseases.
Given a surgically implanted BCI, this patient was able to control his own personal computer independently, the team reported, noting their focus was on “an area [of the brain] canonically associated with speech and orofacial movements.”
The researchers stressed that this technology is in its infancy, and that much work is needed to advance brain-computer interfaces as a clinical tool. Still, the team noted that “we demonstrated a rapidly calibrating, accurate … BCI.”
The study, “Speech motor cortex enables BCI cursor control and click,” was published in the Journal of Neural Engineering.
Emerging technology may help restore function lost to ALS
ALS is marked by the dysfunction and death of nerve cells that control movements, with functions such as walking, using the arms, speaking, and swallowing becoming increasingly harder to do as the disease advances. In the late stages of ALS, patients often lose the ability to move and speak.
An emerging technology known as brain-computer interfaces, or BCIs for short, aims to restore function and give more independence to people who have lost motor abilities due to ALS or other diseases.
The basic idea behind BCIs is to implant a sensor into the brain that can detect brain activity and then translate it into activities like speaking or moving a mouse around a computer screen.
Studies using BCIs that aim to let patients control a mouse on a computer screen generally use sensors that are implanted into parts of the brain that help to control the arms and hands. When patients think about moving a limb, the signals can be translated into certain mouse movements.
By contrast, BCIs designed to detect speech are often implanted into the ventral precentral gyrus, or vPCG, which is a part of the brain that helps control mouth and tongue movements during speech.
Although both of these strategies can theoretically give patients some freedom and independence, this paradigm leads to a potential trade-off where patients need to choose either to be able to use a computer or to have their intended speech decoded, depending on where in the brain sensors are implanted.
Here, researchers wanted to see if they could use sensors in the vPCG to control a computer mouse in addition to decoding speech — which could theoretically give patients more flexibility in communicating and navigating the world.
“As speech BCIs driven by vPCG mature and proliferate, it is valuable to identify what other BCI applications vPCG can enable,” the researchers wrote.
For example, “by controlling a computer cursor with … neural signals, a person with paralysis can type sentences using an on-screen keyboard, send emails and text messages, or use a web browser and many other software applications,” the team wrote.
Testing the use of BCI sensors in different brain regions
The patient here is a participant in BrainGate2 (NCT00912041), a pilot study testing a BCI device in people paralyzed due to ALS or other conditions. The trial is still enrolling participants at sites in the U.S.
This patient had four BCI sensors implanted around the vPCG, and researchers had previously shown that they could use this BCI setup to detect the patient’s speech with high accuracy. Now, the scientists tried to expand the device’s setup to see if they could also use this system to let the patient move a mouse and click on his computer screen. Essentially, rather than trying to speak, the patient would think about moving his head, mouth, or hand in specific ways to move the cursor and click.
A participant with ALS who previously achieved highly accurate speech decoding was able to control his own computer as if using a computer mouse. … This marks an important step toward delivering high-performing, multimodal BCIs to people with paralysis.
The man’s ability to use the computer setup was tested with a standard test involving clicking on specific boxes on the screen. He was 93% accurate in clicking on the right box, and his speed and accuracy were overall on par with what’s been seen with BCI devices implanted into parts of the brain that control arm and hand movements.
“Here, a participant with ALS who previously achieved highly accurate speech decoding was able to control his own computer as if using a computer mouse,” the researchers wrote, noting that “this marks an important step toward delivering high-performing, multimodal BCIs to people with paralysis.”
Experiments were done in which the man tried to use the BCI to speak and use the computer cursor at the same time. This was found to be possible, but the patient’s accuracy with the computer was notably diminished compared with when he was not speaking.
The team noted this as an area for improvement with future BCI devices. In the meantime, the researchers noted that users of this type of BCI would probably have the most success if they spoke and used the computer separately, rather than trying to multitask both at the same time.