AI Method Turns Toe Scan into Rapid PAD Screening Tool

Many cases go undetected because diagnosis often requires a specialized visit for an ankle-brachial index (ABI) test. Missed diagnoses delay therapy and contribute to preventable amputations, particularly in underserved communities. To help address this challenge, researchers have now developed an artificial intelligence (AI) method that uses a seconds-long toe scan to screen for PAD.

Investigators at the University of California San Diego report that photoplethysmography (PPG) combined with AI can function as a rapid, noninvasive digital biomarker for PAD. The approach was developed across the Jacobs School of Engineering, the Design Lab, and UC San Diego School of Medicine, including the Divisions of Cardiovascular Medicine and of Vascular and Endovascular Surgery. The study is published in npj Digital Medicine.

The method uses a light-based PPG sensor on the toe to capture changes in blood volume. A photosensor measures reflected light to derive the PPG signal, which contains physiologic information about peripheral perfusion. The team extracted 78 PPG features that were significantly correlated with ABI results and trained a machine learning model to determine the likelihood of PAD from those features alone.

Researchers assembled more than 10,000 toe PPG recordings from over 3,500 patients who underwent ABI testing at UC San Diego Health between 2020 and 2025. Using only PPG data, the model correctly distinguished PAD cases approximately 83% of the time, compared with roughly 60%–65% performance for traditional clinical risk-factor assessment. Adding smoking status improved the model by another 2%.

Performance was similar across Black, Hispanic, and white patients and across two UC San Diego Health medical campuses. The dataset also included patients with coronary artery disease, diabetes, and end-stage renal disease, supporting generalizability within complex cardiometabolic populations. The researchers state that PPG screening is not intended to replace ABI but could complement it as a triage tool.

Next steps include validating the approach across PPG-capturing devices such as smartphones, pulse oximeters, and wearables, and exploring point-of-care and at-home screening pathways to expedite referrals before major adverse limb events occur.

“An ABI takes 15–30 minutes once the patient is in the clinic, whereas a PPG screen takes only a few seconds,” said Mattheus Ramsis, MD, assistant professor of medicine and medical director of cardiology informatics in the Division of Cardiovascular Medicine at UC San Diego School of Medicine. “Because 95% of people own a smartphone or some advanced device, we can potentially bypass the transportation, financial and institutional barriers that currently limit access to ABI testing."

"If we can catch PAD early enough to prevent a limb amputation, that would be the ultimate impact: preserving limb function, reducing mortality and addressing barriers in underserved populations," said Dr. Ramsis.

Related Links
UC San Diego