Elon Musk's Neuralink Receives FDA Approval for "Blindsight" Brain Implant
Just last week, Elon Musk's brain-computer interface venture, Neuralink, took a significant leap forward. This wasn't merely an upgrade of their existing products, but rather the realization of a bold vision Musk outlined in June: "Blindsight". This new technology is on the verge of receiving approval for clinical trials.
On September 17th, Neuralink announced that Blindsight had been granted "Breakthrough Device Designation" by the U.S. Food and Drug Administration (FDA).
What Makes it a "Breakthrough"?
This designation means Blindsight will undergo an expedited FDA review process, allowing for faster approval, potentially within six months or less.
What Does "Blindsight" Do?
Musk stated that Blindsight could restore vision to those who have lost their eyes or optic nerves. If their visual cortex is intact, it could even enable the congenitally blind to "see" for the first time. He later revealed that Blindsight had already shown promising results in monkey trials.
While the "resolution will be low at first, kind of like early Nintendo game resolution," Musk believes it could eventually surpass normal human vision.
Brain Implants: From Sci-Fi to Reality
The idea of implanting electronic devices into the brain is no longer confined to science fiction. Artificial vision is at the forefront of this technological revolution. Besides Neuralink, research institutions and companies worldwide are conducting experiments to provide vision to the visually impaired through brain implants.
Let's delve into the current mainstream applications of brain-computer interfaces:
Mind-Controlled Prosthetic Arms
Neuralink isn't the first to explore this territory. Using brain-computer interfaces (BCI) to control prosthetic arms, helping paralyzed individuals regain mobility, has been a primary focus of researchers.
In a groundbreaking 2012 study, researchers connected a BCI to a robotic arm, enabling a patient paralyzed for 14 years due to a stroke to regain partial hand function. Witnessing him drink water using his "hand" for the first time was a momentous occasion.
The initial belief was that further research would only need to refine the precision of BCI control and expand the robotic arm's range of motion for the project to be considered a success.
However, the precision of robotic arms controlled by BCIs proved to be a significant challenge. They couldn't match the speed and accuracy of human limbs. This is partly because healthy individuals rely heavily on touch when grasping objects. Losing that sense and relying solely on vision makes even simple tasks like picking up a cup slow and clumsy. This same issue arose when controlling robotic arms via BCIs.
The Breakthrough of Sensory Feedback
Finally, in 2021, scientists achieved a breakthrough. A 28-year-old man, paralyzed for 10 years due to a spinal cord injury, participated in the study.
Researchers implanted two microelectrode arrays with 88 electrodes into the part of his motor cortex that controls arm and hand movement. These arrays measured and analyzed his intended actions, decoding the signals from his brain neurons and transmitting them to the robotic arm, controlling its grasping movements.
Two more microelectrode arrays with 32 electrodes were implanted into the sensory cortex, responsible for hand and finger sensations. Sensors in the robotic arm's hand received pressure signals and transmitted them to the participant's brain through the BCI, simulating the feeling of touch.
The participant underwent nine object-grasping tests with and without sensory feedback, for a total of 108 trials. Without feedback, his total score was 17, with only one perfect score of 3. With feedback, his score jumped to 21, with 15 perfect scores.
The addition of sensory feedback reduced the experiment's completion time by 51.2%, with the median time dropping from 20.9 seconds to 10.2 seconds.
The researchers attributed this improvement to the reduced time spent positioning the hand to grasp objects. Without tactile feedback, the participant needed more time to find the optimal grasping position.
This breakthrough signifies the increasing maturity of brain-controlled robotic arm technology. While current BCIs can help quadriplegics perform daily tasks, future research will aim to refine control and feedback mechanisms, bringing them closer to the capabilities of natural limbs.
Beyond Prosthetic Limbs: Functional Electrical Stimulation
Researchers are also exploring alternative BCI applications for paralysis patients. Instead of controlling external robotic arms, they're using functional electrical stimulation to control the patient's own limbs.
In 2014, a 53-year-old man with a spinal cord injury joined the BrainGate2 clinical trial. Two microelectrode arrays were implanted in the arm region of his motor cortex. The arrays recorded his brain's neural activity, decoding it into commands that stimulated muscle contractions in his arm through electrodes, allowing him to control an arm support.
Before the BCI and functional electrical stimulation system, the patient could only make slight elbow twitches and couldn't move his hand. Afterward, he could voluntarily control his shoulder, elbow, and hand joints, even grabbing a coffee cup and bringing it to his mouth.
These advancements offer paralyzed individuals the possibility of regaining lost function, promoting independence and improving their quality of life.
Visual Prostheses: Restoring Sight
Visual prostheses are another exciting area of development for assisting people with disabilities. Elon Musk's Blindsight aims to tap into this market.
However, the human eye is an incredibly complex organ. Therefore, visual prostheses operate differently than prosthetic limbs.
Understanding Visual Prostheses
When light enters our eyes, it passes through the cornea and lens, reaching the retina at the back of the eye. Cells called photoreceptors convert light into electrical signals, which are then transmitted to the brain via the optic nerve. Our brain interprets these signals as images.
Vision impairment often stems from damage to the retina or optic nerve, disrupting communication between the eye and the brain. Implants bypass these damaged pathways, sending information directly to the brain, theoretically addressing blindness caused by various eye diseases or injuries.
A Patient's Journey with a Visual Prosthesis
One patient who received a visual prosthesis is Berna Gomez Pastrana. At 56, she had lost sight in her left eye due to retinal detachment at 17 and her right eye to an eye disease in 2016, forcing her to live in darkness.
In 2021, she learned about a visual prosthesis trial at the Illinois Institute of Technology in Chicago. The researchers cautioned her that the device was experimental, and her vision might not fully recover. Despite this, Pastrana chose to participate.
Surgeons performed a craniotomy, removing a portion of her skull to implant 25 chips into her brain. These chips, essentially miniaturized stimulators, emitted mild electrical currents. Each chip, about the size of a pencil eraser, contained 16 microelectrodes thinner than a human hair, each individually controllable. Pastrana had a total of 400 implanted electrodes.
A camera mounted on eyeglasses captured her surroundings. The images were processed by specialized software and converted into commands for the implanted chips, stimulating specific electrodes to evoke visual perceptions called "phosphenes" in Pastrana's brain. These appeared as points of light, even though no actual light entered her eyes.
The electrodes were clustered in one region of the visual cortex, limiting Pastrana's phosphenes to the lower left corner of her visual field.
While not a complete restoration of sight, these phosphenes were enough for Pastrana to navigate rooms and perform basic tasks. She could identify a plate from a set of four objects on a table.
The Path to Clearer Vision
In a study at Miguel Hernandez University, participants received implants with only 100 electrodes. Eduardo Fernández, the lead researcher, emphasized that the current goal is not full visual restoration, but rather improving orientation and mobility for visually impaired individuals. He highlighted a volunteer's ability to avoid obstacles while running on a VR treadmill. Fernández hopes that future implants with more electrodes will create more phosphenes, leading to more detailed images.
Xing Chen, an assistant professor of ophthalmology at the University of Pittsburgh, agrees, believing that restoring sight would require hundreds to thousands of electrodes.
Philip Troyk, the lead researcher of Pastrana's trial, believes that electrode placement is more critical than the number of electrodes. Distributing them more widely across the visual cortex would generate more phosphenes, although this would require more invasive surgery.
Thanks to the 25 chips, Pastrana can now perceive the world through white and iridescent dots. She describes her vision as flashes on a radar screen.
Brain-Computer Interfaces for Communication: Giving Voice to the Voiceless
Besides restoring movement and vision, BCIs can translate thoughts into speech, helping patients with speech impairments communicate.
Casey Harrell, diagnosed with amyotrophic lateral sclerosis (ALS) at 40, experienced progressive paralysis and severe speech difficulties over the next five years. His speech became difficult to understand, even for those closest to him. Frustrated by his inability to communicate effectively, especially with his daughter, Harrell decided to undergo BCI surgery.
Regaining the Ability to Communicate
In July 2023, Harrell enrolled in the BrainGate2 clinical trial. Surgeons made a 5x5 centimeter opening in his skull and implanted four microelectrode arrays, each with 64 electrodes, 1.5 millimeters deep into the left side of his brain's motor cortex, an area responsible for speech production. The surgery lasted five hours.
Three days later, Harrell used the BCI for the first time. After a 30-minute calibration session repeating pre-set sentences containing 50 words, he tried constructing sentences freely. The BCI read the commands his brain sent to his muscles, converting them into words displayed on a screen and read aloud, achieving a remarkable 99.6% accuracy.
The second time, the vocabulary expanded to 125,000 words. After 1.4 hours of training, accuracy reached 90.2%. With further calibration, the word recognition accuracy remained at 97.5% eight months after surgery, with a speed of 31.6 words per minute – about half the speed of normal speech. This far surpassed his communication rate through his own speech or using a head-controlled mouse.
Harrell uses the BCI daily to chat with family, attend work meetings, send emails, and browse the internet. His loved ones say the system's voice resembles his own.
This study demonstrated the potential of BCIs to accurately decode brain activity into speech, offering hope for individuals with severe speech impairments.
The Future of Brain-Computer Interfaces
From the discovery of electroencephalography to today's ability to decode neural signals and control devices, brain-computer interface technology has moved from science fiction to reality.
BCIs hold immense promise, offering not only greater independence for people with disabilities but also potentially enhancing the capabilities of able-bodied individuals.
However, challenges remain. Technological advancements are needed to improve accuracy, stability, and minimize risks. Ethical and legal considerations surrounding privacy, data security, and potential misuse must be addressed.
Despite these challenges, brain-computer interfaces are a source of excitement and hope, with the potential to revolutionize our understanding of the brain and consciousness, and transform how we interact with the world.
What are your thoughts on the future of brain-computer interface technology?