人工心脏输出流量和压力的神经网络估算法  

A METHOD FOR COMPUTING OUTPUT FLOW RATE AND PRESSURE OF ARTIFICIAL HEARTS BY NEURAL NETWORKS

作　　者：封志刚[1] 曾培[1] 茹伟民[1] 袁海宇[1] 李岚[1] 钱坤喜[1] 

机构地区：[1]江苏理工大学生物医学工程研究所,镇江212013

出　　处：《中国生物医学工程学报》2002年第6期568-572,共5页Chinese Journal of Biomedical Engineering

基　　金：国家"九五"攻关项目基金 ( 96- 92 0 - 12 - 0 5 ) ;国家教委博士点基金 ( 976 2 990 3);江苏省科委基金资助项目 (BE96 0 5 6)

摘　　要：人工心脏的输出流量和压力是血泵设计及运行的重要特性参数 ,其测量精度和方法直接关系到人工心脏在动物实验及临床中的实际应用效果。本文提出了一种新的测量方法 ,即用神经网络从叶轮式人工心脏电机驱动参数换算血泵的流量和压力。与传统的测量方法相比 ,本方法具有如下特点 :(1)无创性 ,不会对血液造成破坏 ,同时减少了感染机会。 (2 )结构简单 ,省去了流量计和压力计 ,便于人工心脏完全植入体内。Output flow rate and pressure are very important characteristic parameters of artificial hearts, so the precision and means to measure them have direct effect on animal and clinic performance. On the basis of the experiment with blood pump, the authors put forward a new measuring method in which neural networks were applied to convert drive parameters to flow rate and pressure of the blood pump. In comparison with the traditional measuring method, the advantages of this method are: first, it is non-invasive and does not destroy blood corpuscles and decreases the chances of infection; second, the structure is simplified by leaving out flowmeters and manometers, and thus smoothens the way of implantation of the artificial heart.

关 键 词：输出流量 压力 神经网络估算法 人工心脏 血泵 人工神经网络 叶轮式人工心脏 

分 类 号：R318.11[医药卫生—生物医学工程]