Brain-Computer Interfaces in Healthcare: Improving Communication for the Voiceless

By Maria Fujisawa

In the near future, locked-in patients may regain the ability to communicate through the rapid development of neurotechnologies and brain-computer interfaces. Brain-computer interfaces are opening up new possibilities for people with disabilities, with its potential to restore communication for people who have lost the ability to communicate because of paralysis, strokes, or other neurological conditions.

Understanding Brain-Computer Interfaces

To put it in simple terms, brain-computer interfaces are a way for people to communicate or control external devices through their thoughts. Even though they are a highly experimental still developing technology, they have many different applications including:

• Assistive communication: BCIs can provide a lifeline for people with communication impairments such as ALS or locked in syndrome communicate through their thoughts.
• Neurogaming: BCIs can also offer a new dimension of gaming where the player can control characters and play games using their brain signals
• Motor neuroprosthetics: BCIs can help people with injuries or disabilities regain mobility through controlling prosthetic limbs as well as working as devices such as mind controlled wheelchairs. Some military officials are using motor neuroprosthetics to figure out a way to help them run faster.
• VR/AR: BCIs can enhance VR/AR user experience by allowing users to control objects just by using their thoughts. With the development of BCIs, we might be able to replace monitors.

Its medical applications include its uses in assistive communication for people with paralysis. One of the most notable examples of assistive communication technologies was employed by Stephen Hawking, who had progressive motor neuron disease. He used different computer programs that used predictive word generating software that allowed him to select words and initially controlled the selection of words using his hand, later his cheek. In later years, he collaborated with Intel to explore alternative methods such as eye tracking and electroencephalography (EEG) signals, even though he unfortunately could not adopt them.

A Closer Look at How Brain-Computer Interfaces Work

There are mainly 3 types of brain-computer interfaces (BCIs): invasive, partially invasive, and non-invasive BCIs.

Invasive BCIs are brain-computer interfaces implanted directly into the gray matter of the brain, which contains neuronal cell bodies and dendrites. Unlike non-invasive BCIs, invasive BCIs can record the activity of single neurons but also bring many ethical issues.

Partially invasive BCIs are implanted in the skull but outside the gray matter/ the brain. They have lower risks of forming scar-tissues in the brain.

Non-invasive BCIs could be something like a portable or wearable brain computer interface. They use devices called electroencephalogram, also known as EEGs, and is the least invasive method. Currently, we rely mostly on EEG(electroencephalogram) which measures and records brain activity.

BCIs work by recording neural signals from the brain and translating them into words and phrases. BCIs capture the electrical signals that are produced by the brain’s neural activity using electrodes on the scalp and these sensors measure the frequency of each spike. Then they undergo signal processing, where machine learning is used to interpret the signals. The interpreted signals are then turned into commands that help the user communicate their thoughts or output their thoughts to an external device. BCIs can enable interactions with assistive devices that can be achieved through different output mechanisms such as text generation or cursor control. Typically signals are carried to the speech muscles, but in cases that the patient’s signals do not reach the speech muscles, the BCIs can directly pick up the signals from the speech cortex of the brain.

How could BCIs help with assistive communication?

One of the most revolutionizing applications of brain-computer interfaces is its applications in assistive communication, especially for individuals with conditions such as locked in syndrome, or ALS, a neurodegenerative disease that affects motor neurons and muscle movement. For individuals with these conditions, BCIs offer a solution that can improve communication for them that create a direct bridge between the brain and an external device. Usually for medical conditions like ALS, patients lose all muscle control and can’t control in the traditional ways but can make eye movements such as blinking. BCIs enable patients with such medical conditions to convey their thoughts in real time.

Locked in syndrome is a similar condition, and many locked in patients can communicate through purposeful eye movements. However for some that are completely paralyzed and cannot move their eyeballs as well, they can be trapped in a conscious state unable to communicate. With brain computer interfaces, communication is more convenient and they can convey words the person is saying silently in their head.

Thoughts into words

So how exactly do BCIs turn your thoughts into words?

In a study from nature, by decoding attempted handwriting movements from neural activity in the motor cortex, the BCI translated the thoughts into text and achieved typing speeds of 90 characters per minute with high accuracy. The study participant’s hand was paralyzed due to a spinal cord injury attempted to handwrite letters in their head and was able to convert this to text with machine learning models with high accuracy.

By thinking about different movements, the brain produces different signals that the machine learning models are able to differentiate. In the same experiment from nature, they were able to produce understandable text as seen below, with small errors. The image shows 2 real time trials that show the text the BCI and machine learning model was able to decode without any training.

Challenges and Ethical Considerations

While BCIs hold great potential for revolutionizing healthcare and assistive communication, they also have many challenges and ethical considerations such as:

• Privacy/data: collecting sensitive data related to the individual’s emotions- With the potential for thoughts to be read and manipulated, there could be ethical issues with maintaining privacy and private thought.
• Access: BCIs should be focused on inclusive access and not widening the differences in healthcare based on an individual’s socioeconomic status.
• Accuracy: Neural signals can vary from person to person/ there may also be environmental factors such as interference with noise which could prevent the BCIs from getting accurate signals.

As of now, brain-computer interfaces are still a highly experimental area of research, it has the potential to revolutionize so many areas of our lives, including the medical, social aspects of our lives. With companies creating more convenient types like eeg headbands, and even ear-brain interfaces, these developments are slowly helping with the integration of bcis in our daily lives.