Related Experiment Video
Updated: Aug 25, 2025
06:34
Evaluation of Left Ventricular Structure and Function using 3D Echocardiography
Published on: October 28, 2020
4.1K
Estimation of Left Ventricular Ejection Fraction Using Cardiovascular Hemodynamic Parameters and Pulse Morphological
Shing-Hong Liu1, Zhi-Kai Yang1, Kuo-Li Pan2,3,4
1Department of Computer Science and Information Engineering, Chaoyang University of Technology, Taichung City 41349, Taiwan.
Nutrients
|October 14, 2022
Summary
This study estimates left ventricular ejection fraction (LVEF) using non-imaging methods. Machine learning models accurately predict LVEF, offering a convenient alternative for heart failure diagnosis.
Area of Science:
- Cardiology
- Biomedical Engineering
- Machine Learning
Background:
- Heart failure (HF) affects many, particularly older adults, requiring continuous management.
- Left ventricular ejection fraction (LVEF) is crucial for diagnosing HF.
- Current LVEF assessment often relies on imaging techniques.
Purpose of the Study:
- To estimate LVEF using cardiovascular hemodynamic parameters, pulse morphology, and bodily information.
- To evaluate the efficacy of machine learning algorithms for LVEF estimation.
- To develop a non-imaging method for convenient LVEF measurement.
Main Methods:
- Utilized self-constructing neural fuzzy inference network (SoNFIN) and XGBoost regression models.
- Employed recursive feature elimination for optimal parameter selection.
- Trained and tested models on data from 20 heart failure patients.
Main Results:
- XGBoost achieved an estimating root-mean-square error (ERMS) of 6.4 ± 2.4%.
- SoNFIN achieved an ERMS of 6.9 ± 2.3%.
- Both models demonstrated accuracy comparable to echocardiography.
Conclusions:
- Machine learning models can accurately estimate LVEF from non-imaging data.
- This approach offers a convenient, non-invasive alternative for LVEF assessment.
- The proposed method holds potential for future clinical application in heart failure management.
