A study by the University of Tokyo and Institute of Science Tokyo has demonstrated that its artificial intelligence (AI)-based algorithm can identify patients with diabetes or hypertension based on short, spectroscopic video recordings of the face.
The pair’s prospective study, the findings of which are set to be presented at the 2026 European Society of Cardiology (ESC) congress, taking place 28-31 August in Munich, recruited 215 participants consisting of patients diagnosed with diabetes and healthy volunteers.
Go deeper with GlobalData
Discover B2B Marketing That Performs
Combine business intelligence and editorial excellence to reach engaged professionals across 36 leading media platforms.
Each participant underwent a short, high-speed video recording of their face and palms using a spectroscopic camera. A machine-learning algorithm subsequently analysed the individual videos and extracted data on pulse-wave dynamics that measured the stiffness of arteries, skin blood-flow patterns and the spectral characteristics of skin colouring.
On the basis of facial blood flow patterns, the algorithm was able to detect diabetes with accuracy at 88.2% from a 30-second video and 81.2% from a five-second video.
As published previously, the algorithm also detected hypertension with 95% accuracy from a 30-second recording. Meanwhile, the sensitivity to detect normal blood pressure was 100%, while hypertension sensitivity was 89.2%. In addition, the algorithm’s accuracy on the basis of a five-second video remained high at 90.3%.
Ryoko Uchida, a researcher within the department of advanced cardiology at the University of Tokyo who is set to present the study’s findings at this year’s ESC congress, commented: “Our machine-learning algorithm accurately detected hypertension and diabetes from facial spectroscopic video recordings as short as five seconds.
“We intend to validate these findings in larger cohorts across more diverse populations to support real-world application. If validated, this contactless approach could allow people to be screened in everyday settings – without cuffs, blood sampling or a dedicated clinic visit – helping to identify at-risk individuals who would otherwise remain undiagnosed and therefore untreated.”
AI is having a significant impact across healthcare, with a pronounced role in the medical imaging space. A report by GlobalData forecasts that AI in healthcare will reach a valuation of $57.4bn in 2029.
Regarding AI’s application in visual imaging, other companies are using the technology to assess eye scans to detect disease. US Food and Drug Administration (FDA)-cleared in July 2026, iHealthScreen has developed a software that uses standard optometry tools to screen patients for diabetic retinopathy. Meanwhile, a research team at the University of Edinburgh is currently developing an AI software for use by optometrists to detect dementia risk from routine eye tests.
