iPhone camera model screened for healthy aging of arteries
Against clinical criteria, iPhone camera recordings classified healthy vascular aging with 70% overall accuracy in an independent group of 64 adults.
Our read
Biomarkers- Studied:
- Cohort study of 64 adults averaging 66.9 years old
- What changed:
- The model identified healthy vascular aging with 70% overall accuracy using iPhone camera waveforms alone
- The catch:
- The independent iPhone test included only 64 adults
- Why it’s aging research:
- It builds or tests a tool for aging research.
Editor of record: Boris Djordjevic · AgingResearch.News
Sixty-four adults averaging 66.9 years old, 36% women, provided iPhone recordings for an independent test of a computer screening model in an observational study. The camera captured wave patterns from the carotid artery in the neck. Researchers developed the model using artery pressure recordings from 5,494 Framingham Heart Study participants, averaging 50.6 years old; 54% were women.
Healthy vascular aging required age 50 or older, blood pressure below 140/90 mm Hg, low artery stiffness and no known use of blood pressure medication. Using camera waveforms alone, the model detected 76% of those who met these criteria and achieved 70% overall accuracy. It needed no blood pressure readings, demographic information or medical history as inputs. Researchers did not retrain or adjust it for the iPhone group.
Why it matters
Preserved artery function is one aspect of healthy aging. The authors suggested that smartphone cameras could make screening more accessible in community and remote settings.
Caveats
The independent iPhone test included only 64 adults; training also counted adults under 50 without diabetes or cardiovascular disease as healthy-aging-like examples. The findings were reported in a conference abstract.
The paper
Screening for Healthy Vascular Aging Using Smartphone Camera-Based Carotid Pressure Waveforms
Univ of Southern California
AHA Scientific Sessions 2026 Abstract, Nov 2, 2026
Presented at AHA Scientific Sessions 2026, abstract 17418
Comments
Be the first to comment.