Development of CV Risk Prediction Tools Based on AI and Fundus Imaging Technology Study (PERFECT)

NCT ID: NCT06181552

Last Updated: 2023-12-26

Basic Information

• Recruitment Status: NOT_YET_RECRUITING
• Total Enrollment: 1072 participants
• Study Classification: OBSERVATIONAL
• Study Start Date: 2023-12-31
• Study Completion Date: 2025-06-30

Brief Summary

This study aims to develop a cardiovascular disease (CVD) screening tool and cardiovascular risk prediction tool based on fundus imaging data with the method of artificial intelligence.

Detailed Description

This study will establish a cohort of individuals including patients with CVD and participants with high CVD risk, and all the study participants will be follow-up for 1 year. By collecting baseline clinical data, fundus imaging data, and CVD events during the follow up, this study aims to distinguish CVD status based on the fundus imaging data, and explore the association between fundus imaging data and occurence of CVD during the follow up. By using machine learning approach, this study aims to construct a CVD screening tool and CVD prediction tool based on fundus imaging data.

Conditions

Cardiovascular Diseases

Keywords

cardiovascular risk prediction; fundus imaging

Study Design

• Observational Model Type: COHORT
• Study Time Perspective: PROSPECTIVE

Study Groups

Participants with CVD

Meeting any of the following:

1. Established coronary heart disease, including previously diagnosed myocardial infarction, previous treatment with coronary intervention or coronary artery bypass grafting, coronary artery stenosis ≥50%, or chest pain with objective evidence of myocardial ischemia (indicated by stress electrocardiogram or stress imaging)

2. Stroke

fundus photograpgy

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo fundus photography.

optical coherence tomography

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo OCT examination.

optical coherence tomography angiography

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo OCT-A examination.

Participants with high CVD risk

Participants without CVD, but meeting at least two of the following:

1. Men aged ≥ 60 years old, or women aged ≥ 65 years old;

2. Diabetes;

3. Total cholesterol>5.2 mmol/L, or LDL-C>3.4 mmol/L, or HDL-C<1.0 mmol/L;

4. Currently smoking, defined as daily smoking lasting for 1 year or more.

fundus photograpgy

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo fundus photography.

optical coherence tomography

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo OCT examination.

optical coherence tomography angiography

Intervention Type: DIAGNOSTIC_TEST

All the participants will undergo OCT-A examination.

Interventions

fundus photograpgy

All the participants will undergo fundus photography.

Intervention Type: DIAGNOSTIC_TEST

optical coherence tomography

All the participants will undergo OCT examination.

Intervention Type: DIAGNOSTIC_TEST

optical coherence tomography angiography

All the participants will undergo OCT-A examination.

Intervention Type: DIAGNOSTIC_TEST

Other Intervention Names

OCT; OCT-A

Eligibility Criteria

Inclusion Criteria

Three types of participants will be included, which are:

• Participants with established coronary heart disease, including previously diagnosed myocardial infarction, previous treatment with coronary intervention or coronary artery bypass grafting, coronary artery stenosis ≥50%, or chest pain with objective evidence of myocardial ischemia (myocardial ischemia indicated by stress electrocardiogram or stress imaging)
• Participants with established stroke.
• Participants without coronary heart disease or stroke, but are at high risk for CVD, defined as meeting at least two of the following:

1. Men aged ≥ 60 years old, or women aged ≥ 65 years old;

2. Diabetes;

3. Total cholesterol>5.2 mmol/L, or LDL-C>3.4 mmol/L, or HDL-C<1.0 mmol/L;

4. Currently smoking, defined as daily smoking lasting for 1 year or more.

Exclusion Criteria

• Participants unable to provide fundus imaging data required for the study due to the following reasons:

1. Permanent blindness, blurred vision, flying mosquito disease, or refractive medium opacity seriously affecting fundus examination, such as severe cataracts, vitreous hemorrhage, etc.

2. Macular edema, severe nonproliferative retinopathy in diabetes, proliferative vitreoretinopathy, radiation ophthalmopathy or retinal vein occlusion

3. Eyeball enucleation, eye deformities, etc.

4. Previous retinal laser therapy, injection therapy for any eye, or history of retinal surgery

5. Photosensitivity, or taking medication that can cause photosensitivity, or currently undergoing photodynamic therapy

6. Unable to cooperate with examination for collection of fundus imaging data

7. Other situations that the participants fail to provide fundus imaging data required for the study

• Suffering from other serious diseases with an expected survival period of less than one year, such as advanced malignant tumors
• Unable to adhere to follow-up
• Other conditions which the researchers consider inappropriate for participants to enroll in the study

• Minimum Eligible Age: 18 Years
• Eligible Sex: ALL
• Accepts Healthy Volunteers: Yes

Sponsors

Beijing Tongren Hospital

OTHER

Sponsor Role: collaborator

China National Center for Cardiovascular Diseases

OTHER_GOV

Sponsor Role: lead

Responsible Party

Responsibility Role: SPONSOR

Principal Investigators

Jing Li, PhD, MD

Role: PRINCIPAL_INVESTIGATOR

National Center for Cardiovascular Diseases, Fuwai Hospital

Central Contacts

Jing Li, PhD, MD

Role: CONTACT

Phone: +86 60866077

Email: jing.li@fwoxford.org

Bin Wang, PhD, MD

Role: CONTACT

Phone: +86 60866220

Email: wangbin@fuwai.com

References

Poplin R, Varadarajan AV, Blumer K, Liu Y, McConnell MV, Corrado GS, Peng L, Webster DR. Prediction of cardiovascular risk factors from retinal fundus photographs via deep learning. Nat Biomed Eng. 2018 Mar;2(3):158-164. doi: 10.1038/s41551-018-0195-0. Epub 2018 Feb 19.

Reference Type: RESULT

PMID: 31015713 (View on PubMed)

Other Identifiers

2023-GSP-GG-10

Identifier Type: -

Identifier Source: org_study_id