Noninvasive predictor of HeartMate XVE pump failure by neural network and waveform analysis

Artificial Deep Neural Network for Sensorless Pump Flow and Hemodynamics Estimation During Continuous-Flow Mechanical Circulatory Support

The objective of this study was to compare the estimates of pump flow and systemic vascular resistance (SVR) derived from a mathematical regression model to those from an artificial deep neural network (ADNN). Hemodynamic and pump-related data were generated using both the Cleveland Clinic continuous-flow total artificial heart (CFTAH) and pediatric CFTAH on a mock circulatory loop. An ADNN was trained with generated data, and a mathematical regression model was also generated using the same data. Finally, the absolute error for the actual measured data and each set of estimated data were compared. A strong correlation was observed between the measured flow and the estimated flow using either method (mathematical, R = 0.97, p < 0.01; ADNN, R = 0.99, p < 0.01). The absolute error was smaller in the ADNN estimation (mathematical, 0.3 L/min; ADNN 0.12 L/min; p < 0.01). Furthermore, strong correlation was observed between measured and estimated SVR (mathematical, R = 0.97, p < 0.01; ADNN, R = 0.99, p < 0.01). The absolute error for ADNN estimation was also smaller than that of the mathematical estimation (mathematical, 463 dynes·sec·cm -5 ; ADNN, 123 dynes·sec·cm -5 , p < 0.01). Therefore, in this study, ADNN estimation was more accurate than mathematical regression estimation. http://links.lww.com/ASAIO/A991.

The History of Durable Left Ventricular Assist Devices and Comparison of Outcomes: HeartWare, HeartMate II, HeartMate 3, and the Future of Mechanical Circulatory Support

The utilization of left ventricular assist devices (LVADs) in end-stage heart failure has doubled in the past ten years and is bound to continue to increase. Since the first of these devices was approved in 1994, the technology has changed tremendously, and so has the medical and surgical management of these patients. In this review, we discuss the history of LVADs, evaluating survival and complications over time. We also aim to discuss practical aspects of the medical and surgical management of LVAD patients and future directions for outcome improvement in this population.

A numerical method to enhance the performance of a cam-type electric motor-driven left ventricular assist device

Pulsatile left ventricular assist devices (LVADs) driven by electric motors have been widely accepted as a treatment of heart failure. Performance enhancement with computer assistance for this kind of LVAD has seldom been reported. In this article, a numerical method is proposed to assist the design of a cam-type pump. The method requires an integrated model of an LVAD system, consisting of a motor, a transmission mechanism, and a cardiovascular circulation. Performance indices, that is, outlet pressure, outlet flow, and pump efficiency, were used to select the best cam profile from six candidates. A prototype pump connected to a mock circulatory loop (MCL) was used to calibrate the friction coefficient of the cam groove and preliminarily evaluate modeling accuracy. In vitro experiments show that the mean outlet pressure and flow can be predicted with high accuracy by the model, and gross geometries of the measurements can also be reproduced. Simulation results demonstrate that as the total peripheral resistance (TPR) is fixed at 1.1 mm Hg.s/mL, the two-cycle 2/3-rise profile is the best. Compared with other profiles, the maximum increases of pressure and flow indices are 75 and 76%, respectively, and the maximum efficiency increase is over 51%. For different TPRs (0.5∼1.5 mm Hg.s/mL) and operation intervals (0.1∼0.4 s) in counterpulsation, the conclusion is also acceptable.