Original Research
Deep Learning-Based Prediction of Right Ventricular Ejection Fraction Using 2D Echocardiograms

https://doi.org/10.1016/j.jcmg.2023.02.017Get rights and content
Under a Creative Commons license
open access

Abstract

Background

Evidence has shown the independent prognostic value of right ventricular (RV) function, even in patients with left-sided heart disease. The most widely used imaging technique to measure RV function is echocardiography; however, conventional 2-dimensional (2D) echocardiographic assessment is unable to leverage the same clinical information that 3-dimensional (3D) echocardiography-derived right ventricular ejection fraction (RVEF) can provide.

Objectives

The authors aimed to implement a deep learning (DL)–based tool to estimate RVEF from 2D echocardiographic videos. In addition, they benchmarked the tool's performance against human expert reading and evaluated the prognostic power of the predicted RVEF values.

Methods

The authors retrospectively identified 831 patients with RVEF measured by 3D echocardiography. All 2D apical 4-chamber view echocardiographic videos of these patients were retrieved (n = 3,583), and each subject was assigned to either the training or the internal validation set (80:20 ratio). Using the videos, several spatiotemporal convolutional neural networks were trained to predict RVEF. The 3 best-performing networks were combined into an ensemble model, which was further evaluated in an external data set containing 1,493 videos of 365 patients with a median follow-up time of 1.9 years.

Results

The ensemble model predicted RVEF with a mean absolute error of 4.57 percentage points in the internal and 5.54 percentage points in the external validation set. In the latter, the model identified RV dysfunction (defined as RVEF <45%) with an accuracy of 78.4%, which was comparable to an expert reader’s visual assessment (77.0%; P = 0.678). The DL-predicted RVEF values were associated with major adverse cardiac events independent of age, sex, and left ventricular systolic function (HR: 0.924 [95% CI: 0.862-0.990]; P = 0.025).

Conclusions

Using 2D echocardiographic videos alone, the proposed DL-based tool can accurately assess RV function, with similar diagnostic and prognostic power as 3D imaging.

Key Words

echocardiography
deep learning
right ventricle
right ventricular dysfunction
right ventricular ejection fraction

Abbreviations and Acronyms

2D
2-dimensional
3D
3-dimensional
AUC
area under the receiver-operating characteristic curve
CMR
cardiac magnetic resonance
DL
deep learning
LV
left ventricular
LVEF
left ventricular ejection fraction
MACE
major adverse cardiac event
MAE
mean absolute error
RV
right ventricular
RVEF
right ventricular ejection fraction

Cited by (0)

The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the Author Center.

Dr Tokodi and Mr Magyar contributed equally to this work as joint first authors.

Drs Horváth, Merkely, and Kovács contributed equally to this work as joint last authors.