| Title: | Non-contact optical blood pressure biometry using AI-based analysis of non-mydriatic fundus imaging |
| Journal: | BMJ Innovations |
| Published: | 4 Feb 2026 |
| DOI: | https://doi.org/10.1136/bmjinnov-2025-001422 |
| URL: | https://innovations.bmj.com/content/bmjinnov/early/2026/02/04/bmjinnov-2025-001422.full.pdf |
Publication 18372
WARNING: the interactive features of this website use CSS3, which your browser does not support. To use the full features of this website, please update your browser.
Abstract
Background/Introduction This study was developed to determine whether a machine learning model could be developed to assess blood pressure with accuracy comparable to arm cuff measurements. Methods A deep learning model was developed based on the UK Biobank dataset and was trained to detect both systolic and diastolic pressure. The hypothesis was formulated after data collection and before the development of the model. A comparison was conducted between arm cuff measurements, as ground truth and results from the model, using mean absolute error (MAE), mean squared error and coefficient of determination (R 2 ). Results Systolic pressure was measured with 9.81 MAE, 165.13 mean squared error and 0.36 R 2 . Diastolic pressure was measured with 6.00 MAE, 58.21 and 0.30 R 2 . Conclusions This model improves on existing research and shows errors comparable to the variability of hand cuff measurements. The use of fundus images to assess blood pressure may be more indicative of long-term hypertension. Additional trials in clinical settings may be necessary, as well as additional prospective studies to validate results. </p>
8 Authors
- Idan Bressler
- Dolev Dollberg
- Rachelle Aviv
- Danny Margalit
- Alon Harris
- Brent Siesky
- Tsontcho Ianchulev
- Zack Dvey-Aharon
1 Application
| Application ID | Title |
|---|---|
| 75750 | Deep learning algorithms based on ocular and health status predict cognitive dysfunction and development of dementia. |
Enabling scientific discoveries that improve human health