Blood Pressure Predicted From Artificial Intelligence Analysis of Retinal Images Correlates With Future Cardiovascular Events
Authors' conclusion
With the variability and challenges of real-world SBP measurement, AI analysis of retinal images may provide a more reliable and accurate biomarker for predicting future ASCVD events than traditionally measured SBP.
Study context
- Population
- UK Biobank individuals without a prior atherosclerotic cardiovascular disease (ASCVD) event before retinal imaging
- Participants
- 51,778 individuals95,665 retinal images
- Exposure
- Artificial intelligence (AI)-predicted systolic blood pressure (SBP) from retinal images vs measured SBP
- Outcome
- Correlation of SBP with future ASCVD events
Methodological detailsdeep-learning model, UK Biobank
- Deep-learning model
- UK Biobank
Main finding
As reported in the abstract
AI-predicted SBP
0.067
correlation
Measured SBP
0.049
correlation
P = 0.008
Observational data: associations, not effects.
Further findings
Overall ASCVD event rate
- Overall ASCVD event rate
- 3.4%
Variability in measured SBP
- Mean absolute difference in measured SBP (mm Hg)
- 8.2
- Participants whose 10-year ASCVD risk score was impacted
- 6%
Original publication
Blood Pressure Predicted From Artificial Intelligence Analysis of Retinal Images Correlates With Future Cardiovascular Events. JACC Adv. Published 18 November 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →