Assessment of Hypertensive Retinopathy Using Convolutional Neural Network "RetinAIcheck"

Assessment of Hypertensive Retinopathy Using Keith Wagener Barker's Classification, Based on Convolutional Neural Network "RetinAIcheck"

The current study is aimed at estimating the diagnostic effectiveness of a developed convolutional neural network (CNN) "RetinAIcheck" in grading the severity of hypertensive retinopathy in patients of the Russian population.

The training data set was obtained from an open source and relabeled by seven independent retina specialists, the sample size was 30,000 fundus photographs. The test sample included 729 patients (1401 eyes) with HR. The reference standard was the result of independent grading of HR stage by two ophthalmologists, controversial clinical cases were evaluated with the involvement of a third ophthalmologist.

Study Overview

• Status: Completed
• Conditions: Hypertensive Retinopathy
• Intervention / Treatment: Diagnostic test: Convolutional neural network "RetinAIcheck"

• Study Type: Observational
• Enrollment (Actual): 729

Contacts and Locations

Study Locations

• Russia
  • Moscow, Russia
    • University Clinical Hospital №1, Sechenov University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Probability Sample

Study Population

The test sample was collected at the Cardiology Clinic of the Sechenov University Clinical Hospital № 1, at the Research Institute of Eye Diseases named after M.M. Krasnov and in the Moscow Regional Clinical Research Institute named after M.F. Vladimirsky (MONIKI).

Description

Inclusion Criteria:

- The presence of a diagnosis of hypertension in the patient's electronic medical record.

Exclusion Criteria:

• anophthalmia,
• optic nerve atrophy,
• eyeball injuries,
• age-related macular degeneration,
• central serous chorioretinopathy,
• central serous chorioretinitis,
• clouding of the optical media of the eye, which affects the quality of the image.

Study Plan

How is the study designed?

Number of groups / cohorts

5

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Class 1 hypertensive retinopathy	Diagnostic test: Convolutional neural network "RetinAIcheck"; A convolutional neural network is a medical decision support system that processes digital fundus photographs obtained during mydriasis and determines the probability of the presence/absence of hypertensive retinopathy and it's grading due to Keith Wagener Barker's classification.
Class 2 hypertensive retinopathy	Diagnostic test: Convolutional neural network "RetinAIcheck"; A convolutional neural network is a medical decision support system that processes digital fundus photographs obtained during mydriasis and determines the probability of the presence/absence of hypertensive retinopathy and it's grading due to Keith Wagener Barker's classification.
Class 3 hypertensive retinopathy	Diagnostic test: Convolutional neural network "RetinAIcheck"; A convolutional neural network is a medical decision support system that processes digital fundus photographs obtained during mydriasis and determines the probability of the presence/absence of hypertensive retinopathy and it's grading due to Keith Wagener Barker's classification.
Class 3+4 hypertensive retinopathy	Diagnostic test: Convolutional neural network "RetinAIcheck"; A convolutional neural network is a medical decision support system that processes digital fundus photographs obtained during mydriasis and determines the probability of the presence/absence of hypertensive retinopathy and it's grading due to Keith Wagener Barker's classification.
Class 0 without signs of hypertensive retinopathy	Diagnostic test: Convolutional neural network "RetinAIcheck"; A convolutional neural network is a medical decision support system that processes digital fundus photographs obtained during mydriasis and determines the probability of the presence/absence of hypertensive retinopathy and it's grading due to Keith Wagener Barker's classification.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Accuracy	The ability of a test to correctly identify the proportion of true positive cases	The ability to correctly identify the presence or absence of condition

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity	The ability of a test to correctly identify the proportion of true positive cases	February 2026
Specificity	The ability of a test to correctly identify the proportion of true negative cases	February 2026

Collaborators and Investigators

• Sponsor: I.M. Sechenov First Moscow State Medical University
• Investigators: Principal Investigator: Philipp Yu Kopylov, Prof., Sechenov First Moscow State Medical University (Sechenov University)

Publications and helpful links

Helpful Links

• Related Info

Study record dates

Study Major Dates

• Study Start (Actual): March 11, 2021
• Primary Completion (Actual): February 26, 2026
• Study Completion (Actual): February 26, 2026

Study Registration Dates

• First Submitted: March 10, 2026
• First Submitted That Met QC Criteria: March 10, 2026
• First Posted (Actual): March 13, 2026

Study Record Updates

• Last Update Posted (Actual): March 16, 2026
• Last Update Submitted That Met QC Criteria: March 13, 2026
• Last Verified: September 1, 2021

More Information

Terms related to this study

• Keywords: hypertensive retinopathy; convolutional neural network
• Additional Relevant MeSH Terms: Vascular Diseases; Cardiovascular Diseases; Eye Diseases; Retinal Diseases; Hypertension; Hypertensive Retinopathy
• Other Study ID Numbers: RetinAIcheck-01

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: Data used in the research project is not openly available, but can be provided upon request to the principal investigator.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No