Mar 16, 2021 5:11 PM

This Startup Wants to Take Your Blood Pressure With an iPhone

“Hey Siri, help me treat my hypertension."

If you buy something using links in our stories, we may earn a commission. This helps support our journalism. Learn more. Please also consider subscribing to WIRED

In 1896, Italian physician Riva Rocci published the first of four papers on an invention that is still widely used. It was his take on the sphygmomanometer, a device to measure the pressure that a pumping heart exerts on the arteries. In other words, blood pressure. Rocci’s basic approach of tying a cuff to the upper arm remains standard, and it is a vital tool because hypertension is one of the most serious medical ailments. The CDC reports that nearly half of all adults in the US have high blood pressure, and it is a primary or contributing factor in 500,000 deaths annually—it’s like Covid-19 every year.

Only a fourth of people with hypertension have it under control, in part because sphygmomanometers, whether used in a doctor’s office or via clunky home units, don’t supply a steady stream of readings, multiple times a day and in different settings, to help determine the proper treatment. A new company, launching today, thinks it can improve those grim statistics. It’s called Riva Health, chosen as a tribute to the Italian inventor. Riva measures blood pressure from the arteries of one’s fingertip, which it captures using the flash of an iPhone’s camera.

Courtesy of Riva Health

Riva’s ambition is to help people with hypertension speedily determine their treatment—not only by taking readings with a phone, but also by employing a team of doctors and artificial intelligence experts to determine the right medicines and dosage, freeing cardiologists from that time-consuming task. This vision flows from the firm’s high-profile CEO, Dag Kittlaus, who previously led the startup called Siri that was purchased by Apple. Kittlaus later left Apple and started Viv, another smart assistant effort that was scooped up by Samsung. But first it will have to prove to the world—and especially the FDA—that its iPhone app’s blood pressure readings are reliable and useful.

The inventor of Riva’s technology is 43-year-old Tuhin Sinha. Born in India, he was raised in Chicago and Wichita. His father had been one of seven siblings, six of whom had died of heart disease before the age of 60. Sinha’s father was among those. Those premature funerals drew Sinha to the health field. He eventually joined UCSF, where he first worked on iPad apps to help handle data in a clinical setting, and he wound up doing business development for the campus. That’s when the head of cardiology of UCSF, Jeffrey Olgin, took him aside to explain that the research unit was getting overwhelmed by both big companies such as Google and tiny startups asking the hospital to try out their technologies. Sinha helped Olgin and others figure out how to tap some of that tech for a huge project known as the Health eHeart Study. Sinha left UCSF and started a company called Vital Labs, where he continued working with Olgin and the university.

Going to the doctor’s office to take blood pressure readings provides an incomplete picture. People are usually nervous in the office, and the readings tilt high. But the bigger obstacle is that with infrequent measurements, doctors can take three to six months to figure out the proper medicine and dosage to control hypertension.

Early on, Olgin and Sinha decided to see how much they could improve outcomes by getting patients to measure blood pressure at home. They undertook a study with home versions of blood pressure cuffs. Every day the participants took their blood pressure. When the study ended, he says, they found the doctors were able to nail down the appropriate treatment in 17 days, on average.

But getting consistent results from the home readings was difficult. Even top-of-the-line home devices sucked—their clunkiness made them hard to use, patients had to remember to keep them charged, and the Bluetooth could be flaky. “It was an absolute nightmare to use those things,” Sinha says. “Everything about the process was painful for these folks.”

A much better alternative would dispense with the unwieldy cuff. As he thought about how he might use a phone instead, Sinha started wondering whether a camera and flash might extract information from the fingertip. “Despite how far the fingertip is from your heart, it has a lot of arteries,” Sinha explains. “So when your heart beats, there’s a huge volume of blood coming through. It’s almost like a sound wave, toggling between expansive and compressive. That’s the blood pulse waveform.”

Sinha’s idea wasn’t original; people have been writing papers about the blood pulse waveform since the 1950s. But nobody had yet found a reliable way to measure blood pressure with a phone. “There are these holy grails in the research community—can you use an ordinary commodity device to capture things?” says Shwetak Patel, a University of Washington professor (and MacArthur Foundation “genius” grant winner) who is involved in developing such at-home testing, most recently for Google Health. Google recently released a Pixel phone app that uses the camera to measure pulse and respiratory rate. But blood pressure, Patel says, is one of those holy grails.

In 2014, one company thought it had an answer. An outfit called Aura released its Instant Blood Pressure app in Apple’s App store. “This app is a breakthrough for blood pressure monitoring” wrote “Archie1986” in the App store’s top review. But when the Federal Trade Commission looked into it, it found that the app didn’t work. And? The FTC also discovered that the gushing endorsement from Archie1986 was posted by Aura’s CEO.

Sinha felt he could do better. Extracting the waveform from a person’s fingertip was the easy part. The tricky part is analyzing it to get a useful blood pressure reading. Sinha says he’s come up with a way, though it still needs to be externally validated.

When Sinha told Olgin about his plans, the cardiologist was curious but cautious. “From a physics standpoint, it made perfect sense. But it wasn't really until I started to see the data come in that I thought that this was really going to work the way it’s supposed to,” he says.

Around the time he was perfecting his tool, Sinha shared his vision with a venture capitalist named Greg Yap whose specialty was health care. When Yap became a partner at Menlo Ventures in 2019, he invested in Sinha’s company and invited him to move into Menlo’s San Francisco office as part of a new project called Menlo Labs.

Working from Menlo Labs, Sinha was still trying to figure out a business model for his idea. He also needed to attract an experienced entrepreneur to lead the company. Another partner at Menlo, Shawn Carolan, had just the person in mind, someone he had funded in multiple successful ventures.

That was Dag Kittlaus. Carolan had invested in Siri and Viv and had become friendly with the former Apple executive; they both had houses in Montana. In the summer of 2020 the pair had lunch. Kittlaus, who had recently survived pancreatic cancer, said he wanted his next venture to focus on digital health. “The one thing I’m missing,” he had told Carolan, “is some amazing technological core innovation that can be a Trojan Horse into a meaningful digital health company.”

Carolan set up a Zoom call between Kittlaus and Sinha. “I hammered him with every possible question, trying to poke a hole in this stuff,” Kittlaus says. Still, he left the call uncertain, and continued the discussion in a marathon follow-up meeting. “I think it lasted about eight hours,” Kittlaus says. “I had to grab my charger in the middle of the call because my full phone had been drained.” (The two still haven’t met in person.)

Kittlaus and Sinha created a plan where Riva would interpret the wave form results for the treating physician, iterating on a patient’s medication and dosage. Because Riva compressed an otherwise painstaking months-long process into a few days, it would justify the price that Riva is setting for its services: $1,500 a year per patient. “You’re like, ‘Wow, that seems massive’,” says Sinha, reading my mind when I heard this. “It is massive. But guess what—if the patient doesn’t have it under control, it costs $2,000 a year. So it’s a massive savings.”

Kittlaus doesn’t want to stop at blood pressure. He and Sinha believe that the wonders of phone sensors—maybe even new ones that get developed specifically to capture biometric signals—will be commonplace. And he envisions similar diagnostic and prescriptive services for a variety of maladies.

Google Health’s Patel, while specifying that he hasn’t seen the data on Riva, agrees that the future of medicine might lie in sophisticated interpretations of the secrets our body shares with our mobile devices. He notes that the measurements we commonly use now—such as blood pressure and heart rate—are actually proxies for more complicated conditions. But we’ve locked ourselves into century-old treatment regimens because we’re still using limited tools to log our medical conditions. With a suite of advanced sensors, we might be able to capture other measurements that are superior to those we use currently.

But that’s for later. Right now Riva is concentrating on getting its system validated and approved. Complicating matters is the arrival of several competitors who are also using the camera lens to capture blood pressure readings. A Swiss initiative—confusingly, also called Riva—was the subject of a pilot trial last year. (The outcome of the study has not been released.) And in January of this year another company announced a beta version of a smartphone app, OptiBP, that measures blood pressure through the lens; a study published in Nature indicates that it works. But Riva (the one I’m writing about in this story) isn’t impressed. The studies they will perform on their Riva project will use a superior means of validation, they say. But they won’t reveal their methods, for competitive reasons. Later this year, Riva plans to conduct a 500 person clinical trial. Then it will run a second, smaller trial and take the results to the FDA.

Whenever the FDA decision comes—and nothing is certain—Sinha and Kittlaus might want to try their own product when they open the email with the result. Either way, the waveform might show a spike. Bad news might require medication and uncomfortable calls to investors. But if the news is good, their bank accounts will spike as well. Fifteen hundred dollars per patient every year. That’s something Riva Rocci would never have imagined.

nproxy.org