Ice Lounge Media

Protected: News

This retina implant lets people with vision loss do a crossword puzzle

IceLoungeMedia IceLoungeMedia

Science Corporation—a competitor to Neuralink founded by the former president of Elon Musk’s brain-interface venture—has leapfrogged its rival after acquiring, at a fire-sale price, a vision implant that’s in advanced testing,.

The implant produces a form of “artificial vision” that lets some patients read text and do crosswords, according to a report published in the New England Journal of Medicine today.

The implant is a microelectronic chip placed under the retina. Using signals from a camera mounted on a pair of glasses, the chip emits bursts of electricity in order to bypass photoreceptor cells damaged by macular degeneration, the leading cause of vision loss in elderly people.

“The magnitude of the effect is what’s notable,” says José-Alain Sahel, a University of Pittsburgh vision scientist who led testing of the system, which is called PRIMA. “There’s a patient in the UK and she is reading the pages of a regular book, which is unprecedented.”  

Until last year, the device was being developed by Pixium Vision, a French startup cofounded by Sahel, which faced bankruptcy after it couldn’t raise more cash.  

That’s when Science Corporation swept in to purchase the company’s assets for about €4 million ($4.7 million), according to court filings.

“Science was able to buy it for very cheap just when the study was coming out, so it was good timing for them,” says Sahel. “They could quickly access very advanced technology that’s closer to the market, which is good for a company to have.”

Science was founded in 2021 by Max Hodak, the first president of Neuralink, after his sudden departure from that company. Since its founding, Science has raised around $290 million, according to the venture capital database Pitchbook, and used the money to launch broad-ranging exploratory research on brain interfaces and new types of vision treatments.

“The ambition here is to build a big, standalone medical technology company that would fit in with an Apple, Samsung, or an Alphabet,” Hodak said in an interview at Science’s labs in Alameda, California in September. “The goal is to change the world in important ways … but we need to make money in order to invest in these programs.”

By acquiring the PRIMA implant program, Science effectively vaulted past years of development and testing. The company has requested approval to sell the eye chip in Europe and is in discussions with regulators in the US.

Unlike Neuralink’s implant, which records brain signals so paralyzed recipients can use their thoughts to move a computer mouse, the retina chip sends information into the brain to produce vision. Because the retina is an outgrowth of the brain, the chip qualifies as a type of brain-computer interface.

Artificial vision systems have been studied for years and one, called the Argus II, even reached the market and was installed in the eyes of about 400 people. But that product was later withdrawn after it proved to be a money-loser, according to Cortigent, the company that now owns that technology.

Thirty-eight patients in Europe received a PRIMA implant in one eye. On average, the study found, they were able to read five additional lines on a vision chart—the kind with rows of letters, each smaller than the last. Some of that improvement was due to what Sahel calls “various tricks” like using a zoom function, which allows patients to zero in on text they want to read.

The type of vision loss being treated with the new implant is called geographic atrophy, in which patients have peripheral vision but can’t make out objects directly in front of them, like words or faces. According to Prevent Blindness, an advocacy organization, this type of central vision loss affects around one in 10 people over 80.  

The implant was originally designed starting 20 years ago by Daniel Palanker, a laser expert and now a professor at Stanford University, who says his breakthrough was realizing that light beams could supply both energy and information to a chip placed under the retina. Other implants, like Argus II, use a wire, which adds complexity.

“The chip has no brains at all. It just turns light into electrical current that flows into the tissue,” says Palanker. “Patients describe the color they see as yellowish blue or sun color.”

The system works using a wearable camera that records a scene and then blasts bright infrared light into the eye, using a wavelength humans can’t see. That light hits the chip, which is covered by “what are basically tiny solar panels,” says Palanker. “We just try to replace the photoreceptors with a photo-array.”

A diagram of how a visual scene could be represented by a retinal implant.
COURTESY SCIENCE CORPORATION

The current system produces about 400 spots of vision, which lets users make out the outlines of words and objects. Palanaker says a next-generation device will have five times as many “pixels” and should let people see more: “What we discovered in the trial is that even though you stimulate individual pixels, patients perceive it as continuous. The patient says ‘I see a line,’ “I see a letter.’”

Palanker says it will be important to keep improving the system because “the market size depends on the quality of the vision produced.”

When Pixium teetered on insolvency, Palanker says, he helped search for a buyer, meeting with Hodak. “It was a fire sale, not a celebration,” he says. “But for me it’s a very lucky outcome, because it means the product is going forward. And the purchase price doesn’t really matter, because there’s a big investment needed to bring it to market. It’s going to cost money.”  

Photo of the PRIMA Glasses and Pocket Processor.
The PRIMA artificial vision system has a battery pack/controller and an eye-mounted camera.
COURTESY SCIENCE CORPORATION

During a visit to Science’s headquarters, Hodak described the company’s effort to redesign the system into something sleeker and more user-friendly. In the original design, in addition to the wearable camera, the patient has to carry around a bulky controller containing a battery and laser, as well as buttons to zoom in and out. 

But Science has already prototyped a version in which those electronics are squeezed into what look like an extra-large pair of sunglasses.

“The implant is great, but we’ll have new glasses on patients fairly shortly,” Hodak says. “This will substantially improve their ability to have it with them all day.” 

Other companies also want to treat blindness with brain-computer interfaces, but some think it might be better to send signals directly into the brain. This year, Neuralink has been touting plans for “Blindsight,” a project to send electrical signals directly into the brain’s visual cortex, bypassing the retina entirely. It has yet to test the approach in a person.

Read more

AI could predict who will have a heart attack

IceLoungeMedia IceLoungeMedia

For all the modern marvels of cardiology, we struggle to predict who will have a heart attack. Many people never get screened at all. Now, startups like Bunkerhill Health, Nanox.AI, and HeartLung Technologies are applying AI algorithms to screen millions of CT scans for early signs of heart disease. This technology could be a breakthrough for public health, applying an old tool to uncover patients whose high risk for a heart attack is hiding in plain sight. But it remains unproven at scale while raising thorny questions about implementation and even how we define disease. 

Last year, an estimated 20 million Americans had chest CT scans done, after an event like a car accident or to screen for lung cancer. Frequently, they show evidence of coronary artery calcium (CAC), a marker for heart attack risk, that is buried or not mentioned in a radiology report focusing on ruling out bony injuries, life-threatening internal trauma, or cancer.

Dedicated testing for CAC remains an underutilized method of predicting heart attack risk. Over decades, plaque in heart arteries moves through its own life cycle, hardening from lipid-rich residue into calcium. Heart attacks themselves typically occur when younger, lipid-rich plaque unpredictably ruptures, kicking off a clotting cascade of inflammation that ultimately blocks the heart’s blood supply. Calcified plaque is generally stable, but finding CAC suggests that younger, more rupture-prone plaque is likely present too. 

Coronary artery calcium can often be spotted on chest CTs, and its concentration can be subjectively described. Normally, quantifying a person’s CAC score involves obtaining a heart-specific CT scan. Algorithms that calculate CAC scores from routine chest CTs, however, could massively expand access to this metric. In practice, these algorithms could then be deployed to alert patients and their doctors about abnormally high scores, encouraging them to seek further care. Today, the footprint of the startups offering AI-derived CAC scores is not large, but it is growing quickly. As their use grows, these algorithms may identify high-risk patients who are traditionally missed or who are on the margins of care. 

Historically, CAC scans were believed to have marginal benefit and were marketed to the worried well. Even today, most insurers won’t cover them. Attitudes, though, may be shifting. More expert groups are endorsing CAC scores as a way to refine cardiovascular risk estimates and persuade skeptical patients to start taking statins. 

The promise of AI-derived CAC scores is part of a broader trend toward mining troves of medical data to spot otherwise undetected disease. But while it seems promising, the practice raises plenty of questions. For example, CAC scores ­haven’t proved useful as a blunt instrument for universal screening. A 2022 Danish study evaluating a population-based program, for example, showed no benefit in mortality rates for patients who had undergone CAC screening tests. If AI delivered this information automatically, would the calculus really shift? 

And with widespread adoption, abnormal CAC scores will become common. Who follows up on these findings? “Many health systems aren’t yet set up to act on incidental calcium findings at scale,” says Nishith Khandwala, the cofounder of Bunkerhill Health. Without a standard procedure for doing so, he says, “you risk creating more work than value.” 

There’s also the question of whether these AI-generated scores would actually improve patient care. For a symptomatic patient, a CAC score of zero may offer false reassurance. For the asymptomatic patient with a high CAC score, the next steps remain uncertain. Beyond statins, it isn’t clear if these patients would benefit from starting costly cholesterol-lowering drugs such as Repatha or other PCSK9-inhibitors. It may encourage some to pursue unnecessary but costly downstream procedures that could even end up doing harm. Currently, AI-derived CAC scoring is not reimbursed as a separate service by Medicare or most insurers. The business case for this technology today, effectively, lies in these potentially perverse incentives. 

At a fundamental level, this approach could actually change how we define disease. Adam Rodman, a hospitalist and AI expert at Beth Israel Deaconess Medical Center in Boston, has observed that AI-derived CAC scores share similarities with the “incidentaloma,” a term coined in the 1980s to describe unexpected findings on CT scans. In both cases, the normal pattern of diagnosis—in which doctors and patients deliberately embark on testing to figure out what’s causing a specific problem—were fundamentally disrupted. But, as Rodman notes, incidentalomas were still found by humans reviewing the scans. 

Now, he says, we are entering an era of “machine-based nosology,” where algorithms define diseases on their own terms. As machines make more diagnoses, they may catch things we miss. But Rodman and I began to wonder if a two-tiered diagnostic future may emerge, where “haves” pay for brand-name algorithms while “have-nots” settle for lesser alternatives. 

For patients who have no risk factors or are detached from regular medical care, an AI-derived CAC score could potentially catch problems earlier and rewrite the script. But how these scores reach people, what is done about them, and whether they can ultimately improve patient outcomes at scale remain open questions. For now—holding the pen as they toggle between patients and algorithmic outputs—clinicians still matter. 

Vishal Khetpal is a fellow in cardiovascular disease. The views expressed in this article do not represent those of his employers. 

Read more

Flowers of the future

IceLoungeMedia IceLoungeMedia

Flowers play a key role in most landscapes, from urban to rural areas. There might be dandelions poking through the cracks in the pavement, wildflowers on the highway median, or poppies covering a hillside. We might notice the time of year they bloom and connect that to our changing climate. Perhaps we are familiar with their cycles: bud, bloom, wilt, seed. Yet flowers have much more to tell in their bright blooms: The very shape they take is formed by local and global climate conditions. 

The form of a flower is a visual display of its climate, if you know what to look for. In a dry year, its petals’ pigmentation may change. In a warm year, the flower might grow bigger. The flower’s ultraviolet-absorbing pigment increases with higher ozone levels. As the climate changes in the future, how might flowers change? 

white flower and a purple flower
Anthocyanins are red or indigo pigments that supply antioxidants and photoprotectants, which help a plant tolerate climate-related stresses such as droughts.
© 2021 SULLIVAN CN, KOSKI MH

An artistic research project called Plant Futures imagines how a single species of flower might evolve in response to climate change between 2023 and 2100—and invites us to reflect on the complex, long-term impacts of our warming world. The project has created one flower for every year from 2023 to 2100. The form of each one is data-driven, based on climate projections and research into how climate influences flowers’ visual attributes. 

two rows of flowers that are both yellow and purple
More ultraviolet pigment protects flowers’ pollen against increasing ozone levels.
MARCO TODESCO
a white flower with a yellow center
Under unpredictable weather conditions, the speculative flowers grow a second layer of petals. In botany, a second layer is called a “double bloom” and arises from random mutations.
COURTESY OF ANNELIE BERNER

Plant Futures began during an artist residency in Helsinki, where I worked closely with the biologist Aku Korhonen to understand how climate change affected the local ecosystem. While exploring the primeval Haltiala forest, I learned of the Circaea alpina, a tiny flower that was once rare in that area but has become more common as temperatures have risen in recent years. Yet its habitat is delicate: The plant requires shade and a moist environment, and the spruce population that provides those conditions is declining in the face of new forest pathogens. I wondered: What if the Circaea alpina could survive in spite of climate uncertainty? If the dark, shaded bogs turn into bright meadows and the wet ground dries out, how might the flower adapt in order to survive? This flower’s potential became the project’s grounding point. 

""
The author studying historical Circaea samples in the Luomus Botanical Collections.
COURTESY OF ANNELIE BERNER

Outside the forest, I worked with botanical experts in the Luomus Botanical Collections. I studied samples of Circaea flowers from as far back as 1906, and I researched historical climate conditions in an attempt to understand how flower size and color related to a year’s temperature and precipitation patterns. 

I researched how other flowering plants respond to changes to their climate conditions and wondered how the Circaea would need to adapt to thrive in a future world. If such changes happened, what would the Circaea look like in 2100? 

""
We designed the future flowers through a combination of data-driven algorithmic mapping and artistic control. I worked with the data artist Marcin Ignac from Variable Studio to create 3D flowers whose appearance was connected to climate data. Using Nodes.io, we made a 3D model of the Circaea alpina based on its current morphology and then mapped how those physical parameters might shift as the climate changes. For example, as the temperature rises and precipitation decreases in the data set, the petal color shifts toward red, reflecting how flowers protect themselves with an increase in anthocyanins. Changes in temperature, carbon dioxide levels, and precipitation rates combine to affect the flowers’ size, density of veins, UV pigments, color, and tendency toward double bloom.
2025: Circaea alpina is ever so slightly larger than usual owing to a warmer summer, but it is otherwise close to the typical Circaea flower in size, color, and other attributes.
2064: We see a bigger flower with more petals, given an increase in carbon dioxide levels and temperature. The bull’s-eye pattern, composed of UV pigment, is bigger and messier because of an increase in ozone and solar radiation. A second tier of petals reflects uncertainty in the climate model.
2074: The flower becomes pinker, an antioxidative response to the stress of consecutive dry days and higher temperatures. Its size increases, primarily because of higher levels of carbon dioxide. The double bloom of petals persists as the climate model’s projections increase in uncertainty.
2100: The flower’s veins are densely packed, which could signal appropriation of a technique leaves use to improve water transport during droughts. It could also be part of a strategy to attract pollinators in the face of worsening air quality that degrades the transmission of scents.
2023—2100: Each year, the speculative flower changes. Size, color, and form shift in accordance with the increased temperature and carbon dioxide levels and the changes in precipitation patterns.
""
In this 10-centimeter cube of plexiglass, the future flowers are “preserved,” allowing the viewer to see them in a comparative, layered view.
COURTESY OF ANNELIE BERNER

Based in Copenhagen, Annelie Berner is a designer, researcher, teacher, and artist specializing in data visualization.

Read more
1 … 474 475 476 477 478 … 3,360