Development of an AI Model for Detecting Heart Failure from Fundus Images

Development of an AI Model for Detecting Heart Failure from Fundus Images - Development of an AI Model for Detecting Heart Failure from Fundus Images

Status

Recruiting

Phases

Phase 1

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000058921

Enrollment

400

Registered

2025-12-31

Start date

2025-04-01

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Heart failure

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: (1) Adults aged 20 years or older (2)Patients with cardiovascular disease who are outpatients or inpatients at the Department of Cardiovascular Medicine, The University of Tokyo Hospital, or Moriyama Memorial Hospital (3)Employees of the Meiji Yasuda Health Insurance Association who undergo health checkups Participants who meet criterion (1) and either (2) or (3), and who provide informed consent to participate in this study, will be included.

Exclusion criteria

Exclusion criteria: Patients who do not provide informed consent to participate in the study

Design outcomes

Primary

Measure	Time frame
Accuracy of an AI model for predicting lifestyle-related diseases and elevated NT-proBNP, and for estimating cardiovascular disease (sensitivity, specificity, positive predictive value, negative predictive value)	—

Secondary

Measure	Time frame
Correlation between AI-derived risk scores and outcomes (mortality and incidence of cardiovascular events)	—

Countries

Japan

Contacts

Public ContactEriko Hasumi

The University of Tokyo Hospital Department of cardiology

ehasumi@gmail.com0338155411

Outcome results