A retinal-based AI diagnostic solution that autonomously assesses the future risk of cardiovascular disease (CVD).

See all your CVD risk for having a cardiovascular event using a retinal image


The first retinal-based AI solution to assess your cardiovascular risk

Reti-Intelligence expands solutions to detect heart disease. You can predict your personalized CVD risk without a blood sampling or radiation

Effective CVD risk stratification and triage tool

Our deep-learning algorithm can estimate your Cardiovascular Disease risk as an adjunct to the formal risk assessment tools such as heart CT scan

Meeting prognostic performance of heart CT in projecting CVD events

Reti-CVD can autonomously predict your future CVD risk while meeting the heart CT performance

Meet Reti-CVD

*DrNoon is an approved product name of Reti-Eye & Reti-CVD in some territories

Reti-CVD vs. Heart CT scan for Risk Stratification

When compared with the coronary artery calcium score from the traditional heart CT scan, Reti-CVD successfully stratifies the cumulative cardiovascular disease events risk into three groups(low, moderate, high) and shows similar stratification patterns as heart CT scan

Cardiovascular Disease Events in 5 Years


Retina-based Coronary Artery Calcium score

Heart CT scan

Heart CT-based Coronary Artery Calcium Score

Kaplan-Meier estimates for cumulative incidence of cardiovascular events in the CMERC-HI cohort (A) Cardiovascular disease event by RetiCAC score. (B) Cardiovascular disease event by CAC score. Cumulative event rates of cardiovascular disease events including incident heart failure, stroke, myocardial infarction, and all-cause mortality using the new three-strata risk stratification system of deep learning-based RetiCAC score (A), and CAC stratification (0, >0–100, and >100) (B) in the clinical cohort of CMERC-HI (n=527). CMERC-HI=Cardiovascular and Metabolic Disease Etiology Research Center-High Risk. CAC=coronary artery calcium. RetiCAC=deep-learning retinal coronary artery calcium.

Performance of Reti-CVD vs. Heart CT scan

Reti-CVD and Heart CT scan show the same concordance index of 0.71 based on the pivotal trial. Over 5 years, 28 CVD events occurred in the high-risk group by Reti-CVD, and 27 CVD events occurred in the high-risk group (CAC score >100) by Heart CT scan, confirming similar risk stratification performance. Also only 7 CVD cases occurred in the Reti-CVD low-risk group, but 28 CVD cases occurred in the Reti-CVD high-risk group, indicating 4.10 times higher hazard in the high-risk group compared to the low-risk group from Reti-CVD. This difference was statistically significant with a p-value of 0.023.

Risk of Cardiovascular Disease Events

Reti-CVD Score
Reti-CVD Score
Reti-CVD Score
Reti-CVD Score
Heart CT Scan CAC Score*
Heart CT Scan CAC Score*
Heart CT Scan CAC Score*
Heart CT Scan CAC Score*
Low risk
Moderate Risk
High Risk
0 to 100
over 100
Incidence of CVD**
7 CVD events
11 CVD events
28 CVD events
7 CVD events
15 CVD events
27 CVD events
Hazard Ratio
1 reference
1 reference

*Coronary Artery Calcium score (CAC Score)
**Cases / 100 persons for 10 years

Risk-adjusted model controlling for age, sex, hypertension, dyslipidemia, diabetes, and smoking
Note: Data from CMERC-HI (Cardiovascular and Metabolic Disease Etiology Research Center-High Risk Cohort), N=527



Skip any blood tests or radiation exposure while meeting the CT performance


Save costs through early detection of stroke and heart disease prevention


Provide patients with comprehensive cardiovascular risk management in a primary care setting

Reti CVD vs. current measurement


Measured by Retina-based Coronary Artery Calcium Score

  • No radiation risk or blood test
  • Similar to retina photo cost
  • Available in primary care setting
  • From test to result within a minute
  • Comparable performance to heart CT scan in risk assessment

Heart CT Scan

Measured by Heart CT-based Coronary Artery Calcium Score

  • Exposure to high-dose radiation
  • Not available in primary care settings
  • Relatively high cost
  • Test is taken by operation staff, read by radiologists, and delivered to patients by cardiologists, taking few weeks from test to result

Carotid Ultrasound

Measured by Carotid Intima-Media Thickness

  • Test can be taken only by physicians for over 10 minutes
  • Relatively low accuracy compared to coronary artery calcium score in predicting cardiovascular disease risk
  • Higher cost than taking a retina photo 

How Reti-CVD works

Eye Scan

Take 1 retinal photograph per eye using a fundus camera by optometrists or technicians


Submit images to the cloud for analysis and type patient information


Automatically analyze for signs of CVD risk within a minute


Generate personalized health screening results and download the report


Refer to specialists for further follow-up or suggest lifestyle changes and regular check-ups


Get your result within a minute

Approval of Medical device & Commercial use

Approved in 8 territories, Reimbursed in Korea

Mediwhale is the only company that obtained approval for cardiovascular disease risk assessment using retinal images as a biomarker. Reti-CVD has approvals for medical devices in the EU and Asia and is under FDA clearance in the US.

Reti-CVD has been selected for the Postponement of New Health Technology Assessment (nHTA) in Korea. It has become the pioneering medical AI technology that physicians can use for early detection of CVD. Patients now have a formal process to get reimbursement for Reti-CVD. 

DrNoon is an approved product name of Reti-Eye & Reti-CVD in some territories.


Deep-learning-based cardiovascular risk stratification using coronary artery calcium scores predicted from retinal photographs

The Lancet Digital Health

Coronary Artery Calcium (CAC) score is a clinically validated marker of cardiovascular disease risk. 


Pivotal trial of a deep-learning-based retinal biomarker (Reti-CVD) in the prediction of cardiovascular disease: data from CMERC-HI


The potential of using retinal images as a biomarker of cardiovascular disease (CVD) risk has gained significant attention, but regulatory approval of such artificial intelligence (AI) algorithms is lacking.