Heart Failure With Improved Ejection Fraction and Deep Learning

Yihui Kong

Status: Completed

Conditions: Diagnosis; Heart Failure

Study type: Observational

Funder types: Other

Identifiers

NCT06070506

2020020668

Details and patient eligibility

About

The aim of this study was to design a deep learning-based trained model to assist in HFimpEF diagnosis.

Enrollment: 422 patients

Sex: All

Ages: 18+ years old

Volunteers: No Healthy Volunteers

Inclusion criteria

1. Age >18 years.
2. The diagnostic criteria of HF follows the 2018 Chinese Guidelines for the Diagnosis and Treatment of Heart Failure, having symptoms of dyspnea, fatigue or decreased activity tolerance, having signs of fluid retention (such as pulmonary congestion and peripheral edema), having echocardiogram abnormalities in cardiac structure and/or function, showing elevated natriuretic peptide levels (BNP>35 ng/L or/and N-terminal pro-BNP >125 ng/L).
3. Have reviewing echocardiography after discharge.

Exclusion criteria

1. Patients with hypertrophic, restrictive, or invasive cardiomyopathy and congenital or rheumatic heart disease.
2. Patients with heart transplantation during follow-up.

Trial design

422 participants in 2 patient groups

HFrEF group

Description:

Heart failure patients with LVEF persistently ≤40%.

HFimpEF group

Description:

Heart failure patients with previous LVEF ≤40% and a follow-up LVEF of more than 40%.