Computer scientist Shwetak Patel talks a mile a minute. And there’s a reason for that: he’s got a lot on his mind.
At just 35 years old, Patel is an endowed professor of computer science and electrical engineering at the University of Washington, the founder of several technology startups, the CTO of the UW’s Global Innovation Exchange and a certified genius — he won a prestigious MacArthur Fellowship for his work, joining the ranks of MacArthur “geniuses,” as the fellows are called.
Now, building on years of experience in computer science, Patel is turning his mind to a new challenge: health.
Through his work at the UW and his new startup, Senosis Health, Patel is developing apps for smartphones that can measure health almost as well as expensive medical equipment found in hospitals.
He spoke to GeekWire about his work for this first episode of our new Health Tech Podcast, where we tell the story of digital health innovation and the fantastic minds making it all happen.
Patel said his interest in health is all about using creative engineering skills to make a difference in the lives of people around the world. “I’ve always been an advocate of computing technologies to play roles in these big, grand challenges, and I think the space was such an exciting space for me because I think we could have direct impact on people’s health,” he said.
One huge roadblock in health is access: both physical access to healthcare services and access to the money needed to fund those services. Patel and his fellow researchers have been working to make access to healthcare easier for people around the world.
“Our idea has been: How do we repurpose the sensors that are already on a mobile phone to do similar things that you would find in a clinician’s office or at a hospital? And so, we’re looking at how to use microphones, the camera, the flash, the accelerometer, the gyro in new ways that people never used them before,” he said.
Those apps include BPSense, which uses a smartphone camera and flash together to measure blood pressure.
HemaApp takes that idea even further by using the same sensors to measure the specific light frequencies that bounce through someone’s finger.
“By modeling that, and looking at it through the camera, we can tell you the amount of hemoglobin that’s in your blood,” Patel said. A low hemoglobin count can indicate diseases like anemia or some cancers, and measuring it normally requires taking a blood sample and analyzing it in a lab.
Like the other apps Patel is working on, HemaApp isn’t a full replacement for clinical-grade machines, and can’t diagnose disease on its own. But it can help people decide if they need to seek that more specialized, costly care and help avoid needless expenses for those who don’t need it.
SpiroSmart, another app, uses machine learning to measure the capacity of someone’s lungs, an important measure to diagnose conditions like asthma.
“You hold the phone in front of you, you blow at the face of the phone, and you use this machine learning and signal processing on the audio data to calculate what the flow is,” Patel said. In a clinical trial of the app, Senosis found that it was almost as accurate as a clinical spirometer, which generally costs $10,000, he said.
Some of his projects are even more ambitious. One app, called CoughSense, monitors how often the phone’s microphone hears someone coughing. It can use that data to figure out how diseases are spreading through large populations.
The widespread use of smartphones means these apps can make their way into corners of the world that might not have access to expensive medical equipment, and they could bring down the cost of medical treatment in places like the United States.
Taking such a revolutionary approach to developing medical services isn’t all smooth sailing. Patel recounted meeting with officials from the Food and Drug Administration to talk about how the apps could be approved for use in the U.S.
Patel said they officials didn’t understand what he was trying to do when he showed them the SpiroSmart app. He handed them his phone with the app pulled up, and they asked to see the “device.”
“Well, it’s not a device,” he said. “It’s source code on a phone.” At the time, the FDA officials had never seen anything like it.
“And so, we fast-forward three or four years now, they’re starting to appreciate and understand this,” he said.
As healthcare costs in the U.S. skyrocket, work like Patel’s is becoming more and more accepted and appreciated in the traditionally tech-aversive health industry. He’s one of many innovators working to help people around the world live healthier lives.