ISSN:
1741-0444
Keywords:
Control
;
Least squares
;
Muscle relaxants
;
Neuromuscular blockade
;
Parameter estimation
;
Simulation
;
State estimation
Source:
Springer Online Journal Archives 1860-2000
Topics:
Biology
,
Chemistry and Pharmacology
,
Medicine
Notes:
Abstract An existing algorithm designed to control d-tubocurarine-induced muscle relaxation in humans has been evaluated in 38 clinical trials on patients undergoing upper and lower abdominal surgery. The data captured during these trials were used to perform offline analysis of the algorithm reported on here, by which the correct functioning of the algorthm was established and its efficacy rated. It is shown that, for practical purposes, the average controller performance is close to the theoretical limit which can be achieved for this drug. The use of online parameter estimation makes precise initialisation of the controller less critical than would be the case with a fixed control low. In the trials undertaken, an average time of 12 min was needed to reach the chosen set point and excessive overshoot was avoided in every case. With a setpoint for the single twitch response of 20 per cent of the control height, the average drug consumption for the first hour of operation was 0·38 mg kg−1.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF02455310
