1. Real-time regurgitation estimation in percutaneous left ventricular assist device fully supported condition using an unscented Kalman filter.
- Author
-
Yin, Anyun, Wen, Biyang, Xie, Qilian, and Dai, Ming
- Subjects
- *
HEART assist devices , *ARTIFICIAL blood circulation , *CARDIOVASCULAR system , *STATISTICAL hypothesis testing , *MECHANICAL hearts , *KALMAN filtering , *PULSATILE flow - Abstract
This article discusses a study on the estimation of regurgitation in patients with percutaneous left ventricular assist devices (PLVAD) during full support. The study proposes an estimation system based on a regurgitation model and uses an unscented Kalman filter (UKF) as an estimation approach. The research findings demonstrate that the proposed system can accurately estimate regurgitation during PLVAD full support, providing valuable information for clinical practitioners. The text also discusses the use of the UKF algorithm to estimate regurgitation caused by positional abnormalities after the implantation of a partial left ventricular assist device (PLVAD). The study analyzes experimental data for different degrees of positional offset and finds that the estimation error is minor for moderate and severe offsets, but slightly larger for mild offset. The UKF algorithm is shown to accurately estimate regurgitation in cases of cardiogenic shock. However, further research is needed to improve the accuracy of estimation and validate the results in larger animal models and clinical patient populations. The text also discusses the use of the UKF algorithm to estimate regurgitation status in patients with cardiogenic shock. The algorithm shows relatively accurate estimation, but further consideration is needed to improve accuracy by accounting for the impact of myocardial contractility on pump performance. Real-time observation of flow waveforms can assist in detecting regurgitation events and improving patient safety. The study is limited to a computational approach and a mock circuit, but future developments should incorporate animal and human data for [Extracted from the article]
- Published
- 2024
- Full Text
- View/download PDF