Researchers have effectively bypassed the eyes with a brain implant that permits rudimentary sight.
“Allí,” says Bernardeta Gómez in her local Spanish, aiming to a large black line passing across a white sheet of cardboard upheld at arm’s length in front of her. “There.”
It isn’t exactly an impressive accomplishment for a 57-year-old woman—but that Gómez is blind. And she’s been in the same condition for over a decade. When she was 42, toxic optic neuropathy damaged the bundles of nerves that join Gómez’s eyes to her brain, leaving her totally without sight. She’s unable even to detect light.
But after 16 years of gloom, Gómez was provided a six-month window during which she could see a very low-resolution resemblance of the world represented by glowing white-yellow shapes and dots. This was possible thanks to an adjusted pair of glasses, blacked out and installed with a tiny camera. The apparatus is connected to a computer that processes a live video feed, converting it into electronic signals. A cable overhanging from the ceiling links the system to a port implanted in the back of Gómez’s skull that is connected to a 100-electrode implant in the visual cortex in the back of her brain.
Using this, Gómez recognized letters, ceiling lights, basic shapes written on paper, and people. She also played a simple Pac-Man–like computer game channelled directly into her brain. Four days a week for the period of the experiment, Gómez was guided to a lab by her sighted husband and attached into the system.
Gómez’s first moment of vision, at the end of 2018, was the peak of decades of research by Eduardo Fernandez, manager of neuroengineering at the University of Miguel Hernandez, in Elche, Spain. His objective: to give back sight to as many as possible of the 36 million blind people around the world who desire to see again. Fernandez’s attitude is particularly stimulating because it dodges the eye and optical nerves.
Much former research attempted to restore vision by producing an artificial eye or retina. It succeeded, but the huge majority of blind people, like Gómez, have impairment to the nerve system linking the retina to the rear of the brain. An artificial eye won’t resolve their blindness. For this reason, in 2015, the company Second Sight, which took approval to sell an artificial retina in 2011 in Europe—and in the US in 2013—for an uncommon disease called retinitis pigmentosa, transferred two decades of work away from the retina to the cortex. (Second Sight says somewhat more than 350 people are using its Argus II retinal implant.)
Fernandez believes that progress in implant technology, and a more sophisticated understanding of the human visual system, have given him the self-assurance to go straight to the brain. “The info in the nervous system is the same info that’s in an electrical device,” he says
Repairing sight by feeding signals straight to the brain is ambitious. But the fundamental principles have been used in human-electronic implants in conventional medicine for decades. “Presently,” Fernandez explains, “we have many electric devices cooperating with the human body. For the sensory system, we have a cochlear implant. We also have the pacemaker.”
This former device is the hearing version of the prosthesis Fernandez made for Gómez: an external microphone and processing system that conveys a digital signal to an embed in the inner ear. The implant’s electrodes direct pulses of current into adjoining nerves that the brain reads as sound. The cochlear implant, which was first fitted in a patient in 1961, lets more than half a million people worldwide have conversations as a usual part of everyday life.
“Berna was our first patient, but in the next couple of years we will put in implants in other five blind people,” says Fernandez, who names Gómez by her first name. “We had done same experiments in animals, but a monkey or a cat can’t explain what it’s seeing.”
Her testing took courage. It necessitated brain surgery on an otherwise healthy body—always a dangerous procedure—to fit the implant. And then once more to remove it six months later, since the prosthesis isn’t permitted for longer-term use.
Seizures and phosphenes
Gómez’s voice is the voice of a woman about a decade younger than her age. Her words are restrained, her rhythm is perfectly smooth, and her tone is confident, warm and steady.
When I lastly see her in the lab, I observe Gómez knows the design of the space so well she hardly needs help finding the way in the small hallway and its connected rooms. When I walk over to meet her, Gómez’s face is firstly pointing in the wrong direction ’til I say hi. When I stretch out to shake her hand, her husband directs her hand into mine.
Gómez is at here for a brain MRI to check how things look six months after having her implant detached. She’s also here to see a likely second patient who is in town, and in the room in the course of my visit. At one point during this engagement, as Fernandez clarifies how the hardware connects to the skull, Gómez interjects the discussion, leans forward, and puts the prospect’s hand on the back of her head, where a metal vent used to be. Now there’s virtually no mark of the port. The implant surgery was so ordinary, she says, that she came to the lab the very next day to get plugged in and begin the experiments. She’s had no pain or problems since.
Gómez was fortunate. The extensive history of experiments leading to her fruitful implant has an inconsistent past. In 1929, a German neurologist, Otfrid Foerster found that he could produce a white dot in the vision of a patient if he attached an electrode into the visual cortex of the brain during surgery. He labelled the phenomenon a phosphene. Scientists and sci-fi authors have since visualised the prospective for a camera-to-computer-to-brain visual prosthesis. Some researchers even made rudimentary systems.
In the early 2000s, the theoretical became a truth when an unconventional biomedical researcher named William Dobelle fitted such a prosthesis in the head of a trial patient.
In 2002, the writer Steven Kotler remembered with fear watching Dobelle turn on the electricity and a patient fall to the floor squirming in a seizure. The source was too much stimulation with much current—somewhat, it turns out, brains don’t like. Dobelle’s patients also had complications with infections. Yet Dobelle advertised his bulky device as almost ready for day-to-day use, complete with a publicity video of a visionless man driving slowly and unevenly in a closed parking lot. When Dobelle died in 2004, his prosthesis met the same fate.
Different from Dobelle, who announced a cure for the blind, Fernandez almost continuously says things like, “I don’t want to get any hopes high,” and “We hope to have a method people can use, but right now we’re just carrying early experiments.”
But Gómez did in fact see.
Bed of nails
If the simple idea behind Gómez’s sight—fit a camera into a video cable into the brain—is easy, the details are not. Fernandez and his group first had to work out the camera part. What type of signal does a human retina generate? To attempt to answer this question, Fernandez acquires human retinas from people who have just died, fastens the retinas up to electrodes, reveals them to light, and calculates what hits the electrodes. (His lab has an intimate relationship with the local hospital, which from time to time calls in the middle of the night when a body part donor passes. (A human retina can be kept active for only about seven hours.) His team also uses machine learning to correspond the retina’s electrical output to simple visual inputs, which assist them write software to copy the process automatically.
The following step is taking this signal and sending it to the brain. In the prosthesis Fernandez made for Gómez, a cabled connection goes to a common neuro-implant known as a Utah array, which is only smaller than the elevated tip on the positive end of a AAA battery. Bulging from the implant are 100 tiny electrode spikes, each individual about a millimeter tall—collectively they look like a minute bed of nails. Each electrode can send a current to between one and four neurons. When the implant is put in, the electrodes penetrate the surface of the brain; when it’s detached, 100 small droplets of blood develop in the holes.
Fernandez had to regulate one electrode at a time, sending it progressively strong currents until Gómez perceived where and when she saw a phosphene. Getting all 100 electrodes set in took more than a month.
“The gain to our approach is that the array’s electrodes jut into the brain and sit near to the neurons,” Fernandez says. This enables the implant to generate sight with a much lesser electrical current than was required in Dobelle’s system, which greatly reduces the risk of seizures.
The big disadvantage to the prosthesis—and the main reason Gómez couldn’t keep hers more than six months—is that no one knows how long the electrodes can last without damaging either the implant or the user’s brain. “The body’s defence system begins to break down the electrodes and encircle them with scar tissue, which ultimately weakens the signal,” Fernandez says. There’s also the setback of the electrodes bending as someone moves around. Assessing from research in animals and an initial look at the array Gómez used, he assumes the current arrangement could last two to three years, and maybe up to 10 before it fails. Fernandez expects little minor adjustments will lengthen that to a few decades—a significant prerequisite for a piece of medical hardware that needs invasive brain surgery.
Ultimately, the prosthesis, like a cochlear implant, will need to convey its signal and power wirelessly via the skull to reach the electrodes. But for present, his team has so till now left the prosthesis cabled for experiments— offering the most flexibility to keep upgrading the hardware before confirming on a project.
At 10 pixels by 10 pixels, which is approximately the full potential resolution Gómez’s implant could give, one may recognize basic shapes like a door frame, letters, or a sidewalk. But the outlines of a face, let alone a person, are far more complex. That’s why Fernandez improved his system with image recognition software to recognize a person in a room and emit a pattern of phosphenes to Gómez’s brain that she picked up to recognize.
At 25 by 25 pixels, Fernandez pens in a slide he likes to display, “vision is possible.” And because the Utah array in its existing form is so small and needs so little power to function, Fernandez says there’s no technical reason his team couldn’t fit four to six on each side of the brain, providing vision at 60 x 60 pixels or higher. Yet, nobody knows how much input the human brain can gain from such devices without being stunned and displaying the equal of TV snow.
What it looks like
Gómez told me she would have had the implant installed if she had been given the option and that she’ll be first in line if an upgraded version is available. When Fernandez is done examining her array, Gómez intends to have it framed and hang it on her living room wall.
Back in Fernandez’s lab, he proposes to hook me up to a noninvasive device he uses to examine patients.
Sitting in the same leather chair where Gómez sat during last year’s innovative experiment, I wait as a neurologist grips a wand with two rings touching the side of my head. The tool, called a butterfly coil, is connected to a box that stimulates neurons in the brain with a strong electromagnetic pulse—a sensation called transcranial magnetic stimulation. The first gust feels as if someone is alarming my scalp. My fingers unwillingly curl into my palms. “Look, it functioned!” Fernandez says, chuckling. “That was your motor cortex. Now we will attempt to give you some phosphenes.”
The neurologist relocates the wand and sets the machine for a fast series of pulses. This time when she fires, I feel a powerful zzp-zzp-zzp, as if someone were using the rear of my skull as a door knocker. At that time, although my eyes are wide open, I see something: a vivid horizontal line blazes across the center of my field of vision, together with two shimmering triangles packed with what looks like TV snow. The image disappears as quickly as it came, leaving a short-term afterglow.
“This is similar to what Berna could see,” Fernandez says. But her “sight” of the world was stable on condition that the signal was being communicated to her brain. She could also turn her head with her glasses on, and look around the room. What I had seen were only internal shades of an electrically excited brain. Gómez could genuinely reach out and touch the world she was missing in the last 16 years.