Peripheral artery disease, also known as PAD, is a common but often overlooked condition that affects blood flow in the body.
The disease develops when fatty plaque builds up inside arteries and narrows them over time. This buildup reduces the amount of blood that can travel to the legs and feet.
PAD affects an estimated 8 to 12 million Americans and many more people worldwide. Yet despite being so common, the disease often goes undiagnosed until it becomes severe.
Some patients notice pain while walking or exercising, while others may have numbness, weak legs, slow-healing wounds, or changes in skin color. In many cases, people have no obvious warning signs at all.
This delay in diagnosis can be dangerous. PAD is one of the leading causes of limb amputation because poor blood flow can damage tissue and prevent wounds from healing. The condition is also strongly linked to heart disease and stroke, making it an important public health issue.
Doctors currently diagnose PAD using a procedure called the ankle-brachial index test. During this test, blood pressure measurements from the ankles and arms are compared to check for circulation problems. Although effective, the process is somewhat slow and usually requires special clinic equipment and trained staff.
Researchers at the University of California San Diego believe they may have found a much faster and easier way to screen for PAD. Their new research shows that a technology called photoplethysmography, combined with artificial intelligence, can identify signs of the disease with strong accuracy.
The study was published in the journal npj Digital Medicine.
Photoplethysmography, often shortened to PPG, is already used in many healthcare devices today. It works by shining light into the skin and measuring how much light reflects back. Because blood flow changes the way light behaves inside tissue, the signal can reveal information about circulation.
Most people have unknowingly used PPG before. Pulse oximeters clipped onto fingers in hospitals use this same technology to measure oxygen levels and pulse rate.
In this study, researchers placed the light sensor on patients’ toes to measure blood flow signals. They then used artificial intelligence to study patterns in the signals that may indicate PAD.
The project began after Dr. Mattheus Ramsis, a cardiology informatics specialist at UC San Diego, learned that toe PPG signals were already being collected during standard ABI testing. He realized these signals might hold valuable diagnostic information by themselves.
The research team gathered more than 10,000 toe recordings from over 3,500 patients treated at UC San Diego Health between 2020 and 2025. Using this large dataset, the scientists identified dozens of signal features connected to blood flow problems.
The team then built a machine learning model capable of predicting whether a patient had PAD based only on the PPG signals.
The AI system performed surprisingly well. It correctly distinguished PAD patients from non-PAD patients about 83% of the time. Traditional clinical risk assessment methods usually achieve only about 60% to 65% accuracy.
The researchers also discovered that adding information about smoking history improved the model slightly more. Smoking is one of the strongest known risk factors for PAD because it damages blood vessels and reduces circulation.
Importantly, the AI system worked similarly well across different patient groups. The model performed consistently in Black, Hispanic, and white patients and also worked in people with diabetes, coronary artery disease, and kidney failure.
The researchers believe this approach may eventually make PAD screening much more accessible. Since PPG technology already exists in smartphones, wearables, and pulse-monitoring devices, future versions of the system could potentially allow patients to screen themselves at home.
This could be especially valuable for underserved populations that face transportation, financial, or healthcare access barriers. Earlier detection could allow patients to receive treatment sooner and avoid severe complications such as amputation.
Dr. Ramsis explained that the goal is not to completely replace ABI testing. Instead, PPG screening could serve as a fast triage tool that helps identify patients who need additional evaluation.
The study also reflects a larger trend in medicine toward digital diagnostics. Researchers are increasingly using artificial intelligence to analyze signals from the body, including heart rhythms, breathing patterns, and blood flow data. These digital biomarkers may eventually allow diseases to be detected earlier and more cheaply than traditional testing methods.
In analysing the findings, this research appears exciting because it uses technology that is already widely available. The large patient dataset and strong performance across different racial groups strengthen confidence in the results.
However, the study was conducted within one health system, so more testing in broader populations and on different devices will still be necessary. The accuracy also remains below that of full diagnostic testing, meaning the technology may be best suited as an early screening tool rather than a final diagnosis.
Even with these limitations, the findings suggest that AI-powered blood flow screening could become an important new tool for preventing severe complications from peripheral artery disease.
If you care about heart health, please read studies about how drinking milk affects risks of heart disease , and herbal supplements could harm your heart rhythm.
For more information about heart health, please see recent studies about how espresso coffee affects your cholesterol level, and results showing Vitamin K2 could help reduce heart disease risk.
Source: University of California San Diego.
