A brain-computer system may decipher the thoughts of completely locked-in patients
Such patients have brain function but can't move, speak or blink
"This can be considered to be an important advance," one expert says
You can hear those around you, but you can’t speak. You can feel a touch, but you can’t touch back. You can see, but you can’t move, even to blink your eyes.
That’s the life of a completely locked-in patient, someone who has brain function but complete paralysis, which can be caused by stroke, traumatic brain injury, medication overdose or diseases of the circulatory or nervous system, such as amyotrophic lateral sclerosis (also known as ALS or Lou Gehrig’s disease).
It was generally thought that completely locked-in patients were unable to communicate with the outside world – but a new study has showed otherwise.
An international team of scientists has communicated with completely locked-in patients using a noninvasive brain-computer interface system.
The researchers used the system to decode the patients’ thoughts while the patients were asked yes or no questions, according to the study, published in the journal PLOS Biology on Tuesday.
Niels Birbaumer, a researcher at the Wyss Center for Bio and Neuroengineering in Switzerland and lead author of the new study, said he wasn’t surprised by the findings. “No surprise, but pleasure,” Birbaumer said.
“One patient’s family is using it regularly,” he said of the new system. “With some training, every caretaker of average intelligence can learn it.”
To decode what patients were thinking, the system involved functional near-infrared spectroscopy, a tool that can measure blood flow and oxygenation in the brain, and an electroencephalography (EEG) cap, which can measure electrical activity in the brain.
“Since no other technique allows any communication with these subjects, this can be considered to be an important advance,” said Andrew Schwartz, distinguished professor of neurobiology at the University of Pittsburgh who was not involved in the new study.
‘Are you happy?’
Four patients – a 68-year-old woman, a 76-year-old woman, a 61-year-old man and a 24-year-old woman – who were in completely locked-in states participated in the study.
Over several weeks, the researchers repeatedly asked the patients “yes or no” or “true or false” questions while the patients were connected to the communication system.
As the patients thought about their answers, the researchers analyzed changes in the system’s measurements to determine whether the patient was thinking about a yes or no response.
The mental states for yes or no were different due to oxygenation changes, but the system could not decipher specific letters or words.
The researchers first asked the patients to respond to questions or statements with known answers, such as “You were born in Berlin,” “Your husband’s name is Joachim,” “Paris is the capital of Germany” or “Paris is the capital of France.”
The patients responded correctly at a rate of about 70%, the researchers found.
Once the researchers determined that the patients were trained on how to respond to the questions, they repeatedly asked open questions with no known answers, such as “Are you happy?”
The researchers found that the patients repeatedly answered quality of life questions with a “yes” response, indicating that they had a positive attitude toward life.
A total of at least 200 known questions and 40 open questions were constructed for the patients with input from family members.
Birbaumer and his colleagues had used only near-infrared spectroscopy to communicate with just one completely locked-in patient in 2014 and published their findings in the journal Neurology. He said the new findings mark about 25 years of research.
The new study was successful in showing a creative approach to establishing some communication with completely locked-in ALS patients, said Dr. Brian Litt, a professor and director of the Penn Center for Neuroengineering and Therapeutics at the University of Pennsylvania.
“Their conclusion is, this could be a promising, minimally invasive technology that you could use,” said Litt, who was not involved in the new study.
“I think it’s pretty convincing that they could get ‘yes or no’ answers out of these patients. Now, whether they could get to more sophisticated results, like being able to get language or being able to type, for example, from just your thoughts, that takes more research,” he said. “Keep in mind what the goal would be. The goal would be for somebody to think their thoughts and really give you more meaningful communication.”
Brain implants and blinks
Some researchers have used implantable brain chips or blinking as a way to communicate with locked-in patients, depending on the severity of the patient’s locked-in state.
Though some locked-in patients have some control of their eye movements, those who are completely locked-in have lost such control.
One of the earliest-known records of locked-in syndrome appeared in Alexandre Dumas’ 1844 novel “The Count of Monte Cristo,” in which Noirtier de Villefort suffers a stroke and is unable to move and speak but can control his eyes, exhibiting symptoms of being locked-in (PDF).
Another well-known locked-in case occurred more recently. After French journalist Jean-Dominique Bauby suffered a massive stroke in 1995, he lost the ability to move and speak but could control his eyes, exhibiting a locked-in state. Locked-in syndrome is thought to affect about 1% of stroke victims, although the number could be underestimated.
Before he died in 1997, Bauby authored a memoir about his condition, “The Diving Bell and the Butterfly.” He produced the book by blinking to indicate which letters his transcriber should use to write. The memoir, which was turned into a 2007 film, was written with about 200,000 blinks.
At Northeastern University in Boston, students developed a brain-computer communication device in 2014 that allowed a locked-in patient to control an on-screen keyboard with only his eyes in order to type messages.
The patient would blink to select certain letters on the screen and put them into sentences.
“We still regularly get contacts from people all around the world who ask, ‘Where can I buy this device?’ ” said Waleed Meleis, associate professor and associate chairman of the department of electrical and computer engineering at Northeastern University, who advised students. The device is not on the market for sale or use.
In the past few years, more communication systems intended for locked-in patients have emerged, including the brain-computer interface system demonstrated in the new study. Meleis was impressed by the study’s thoroughness, he said.
“Usually in computer science, you train a computer to do something using machine learning. But here, they are training a person who cannot communicate to manipulate their brain states to produce an answer,” Meleis said of the authors of the new study, with which he was not involved.
“They’re tackling a problem that is real and very challenging,” he said.
Meleis thinks it will still take time before any such communication device may be commonly used with locked-in patients. First, research needs to be replicated and conducted in real-world settings, he said.
“There is a long way to go,” he said. “Many of these users are not in hospitals; they are in nursing homes or care facilities. So I think there’s another leap to be made to adapt these processes to the lifestyles of a user, but I think we’re going to get there.”
Additionally, more research is needed to ensure that questions can be asked to patients with a higher accuracy, as in the new study, patients answered known questions with only a 70% correct response rate.
‘This would give patients that power’
The extent to which brain-computer technology might be used with completely locked-in patients may depend on how easily and economically the technology can be transformed for real-world settings with everyday caregivers, said Gerwin Schalk, associate professor at the University at Albany School of Public Health and research scientist at the National Center for Adaptive Neurotechnologies.
“The biggest barriers to widespread use are probably the economics of the substantial cost of developing and maintaining the system, combined with the fact it only really helps very few people, so it’s sort of like an orphan drug. … It’s very helpful but only to very few people,” said Schalk, who was not involved in the new study.
“There are also important ethical implications,” he added. For instance, doctors might recommend that a patient’s life support be terminated, but then a brain-computer interface might allow that patient to communicate that he or she still wants to live. What should happen next?
Follow CNN Health on Facebook and Twitter
Such brain-computer interfaces can put the power to make decisions about their own care back into patients’ control, said Dr. Karunesh Ganguly, associate professor of neurology at the University of California, San Francisco and a staff neurologist at the San Francisco Veterans Affair Medical Center.
“It certainly allows the possibility for this noninvasive communication to have important implications for these types of patients’ care,” Ganguly said of the new study, with which he was not involved.
“Often, for example, for medical care, you need to be able to gain consent,” he said. “This would give patients that power to make choices about their own care.”