There is a disease affecting millions of people that most of them have never heard of. Peripheral artery disease, PAD, kills more people than colon cancer and breast cancer combined. Seventy percent of the population has no idea it exists. By the time most patients find out they have it, they are already in a vascular surgeon’s office facing major intervention, or worse, looking at an amputation they never saw coming.
That is the part that keeps me up at night. Not that the disease is hard to treat. It is that we are catching it so late.
This week on The Heart of Innovation, Dr. John Phillips and I sat down with a research team from UC San Diego that is working to change that: vascular surgeon Dr. Elsie Ross, cardiologist Dr. Mattheus Ramsis, and engineering PhD student Ava Faccetti. What they are building together started with a cup of black coffee and a question that nobody had thought to ask before.
Dr. Ross had been in AI since 2013, long before it was a buzzword in medicine. She was thinking about PAD patients showing up too late, primary care doctors not checking pulses, people going years without a diagnosis and then landing in surgery. Dr. Ramsis had been working with PPG signals since a fellowship at UCSF, where he helped co-develop the AFib detection algorithm that Samsung uses in their smartwatch today. When they both ended up at UCSD as medical directors of informatics, Dr. Ramsis pitched his work at a meeting, and Dr. Ross was in the audience. She knocked on his door the next day and said they should talk.
Over coffee, she told him about PAD and the physiology behind it. He told her about PPG signals. And then somebody mentioned that a PPG signal is captured at the same time a patient gets an ankle-brachial index test. They looked at each other and both had the same thought at the same moment: there is probably something here.
PPG stands for photoplethysmography, which is why everyone on the show quickly agreed to just call it PPG. It is the same light-reflection technology your smartwatch uses to track your heart rate. You can capture it from any vascular bed, a finger, a wrist, a toe. What Dr. Ross realized was that if you put the sensor at the toe, you are reading the signal right where PAD shows up first. Arteries narrow from the feet up. The toe is the canary.
The team pulled thousands of signals from the vascular lab, with help from medical student collaborator Mustafa who physically extracted them from the records. Ava took those signals and did something that Gemini will describe as AI and that Ava herself would push back on slightly. They did not just feed the data into a model and hope for the best. They sat in meetings with Dr. Ross and looked at signals side by side, some from patients with PAD and some without, and asked her to tell them which was which and why. They were trying to translate physiology into something a computer could reason about. That is a different thing than a black box algorithm spitting out a prediction with no explanation behind it.
They validated the model against the ABI, the ankle-brachial index, which is the clinical gold standard for PAD diagnosis. And the results were strong enough that about nine or ten months ago, sitting in a meeting, they looked at what they had found and said: yeah, this is going to work.
The vision from here is not another expensive machine bolted to the floor of a specialty clinic. It is a consumer device. Something you can use at home, or at CVS, or at your fire department, the same way you check a blood pressure or take a temperature. Not a diagnosis, but an indication, something that says you should probably go see someone about this, before tissue is dying and surgeons are out of options.
Dr. John made the point on air that this is how medicine has to move. We have spent decades building remarkable tools that live inside the walls of specialty practices and academic medical centers. The patients who need them most are the ones least likely to walk through those doors. Rural patients. Patients with no primary care. Asymptomatic patients who feel fine right up until they do not.
An ABI requires you to show up, lie down, have a cuff put on your arm and your leg, and have someone calculate the ratio. That is a real barrier for a lot of people. A four-second toe signal on a phone is not.
I have covered healthcare innovation for a long time and I have sat across from a lot of researchers. What I noticed about this team is that none of them started with the technology and worked backward to find a problem to solve. Dr. Ross started with patients who were losing limbs and worked forward to find a tool that could reach them earlier. It’s important to always start with the patient. It is the difference between a product and a mission.
They are still in the validation phase. There is more work to do. But the core algorithm is built, the foundational paper is complete, and the goal is an FDA-approved screening tool that puts some version of a vascular lab reading in the hands of the people who need it most.
If you want to hear the full conversation, including how Dr. Ramsis accidentally missed the first meeting with Ava’s lab director and it somehow still worked out, click the video above or find us on YouTube at youtube.com/@TheWayToMyHeart or catch us live on AM 1220 KDOW every Saturday morning.
And if you or someone you know has leg pain, wounds that do not heal, or feet that are cold or numb, please do not wait for a device. Call our leg saver hotline at 1-833-PAD-LEGS or visit padhelp.org. The current standard of care can still save a limb if you catch it in time. That is what we are here for.
More on the research:
