文档介绍：IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, VOL. 46, NO. 1, JANUARY 1999 71
Depth of Anesthesia Estimation and Control
Johnnie W. Huang, Ying-Ying Lu, Abinash Nayak, Member, IEEE, and Rob J. Roy* Senior Member, IEEE
Abstract— A fully automated system was developed for the MLAEP Mid-latency auditory evoked potential.
depth of anesthesia estimation and control with the intravenous [Prop] Propofol effect-site concentration.
anesthetic, Propofol. The system determines the anesthesia [Prop] Propofol plasma concentration.
depth by assessing the characteristics of the mid-latency
auditory evoked potentials (MLAEP). The discrete time wavelet [Prop] Propofol setpoint concentration.
transformation was used pacting the MLAEP which ROC Receiver operating curve.
localizes the time and the frequency of the waveform. Feature SDA Step-discriminant analysis.
reduction utilizing step discriminant analysis selected those
wavelet coefficients which best distinguish the waveforms of
those responders from the nonresponders. A total of four I. INTRODUCTION
features chosen by such analysis coupled with the Propofol NTEROPERATIVE awareness due to the inadequate levels
effect-site concentration were used to train a four-layer artificial
work for classifying between the responders and the Iof anesthesia permits explicit cognizance which can often
nonresponders. The Propofol is delivered by a mechanical syringe lead to untoward psychological consequences [1]. A monitor
infusion pump controlled by Stanpump which also estimates capable of estimating DOA is, therefore, desirable for assisting
the Propofol effect-site and plasma concentrations using a the anesthesiologists in minimizing such incidences. However,
partment ic model with the Tackley the significance of the traditional signs of monly
parameter set. In the animal experiments on dogs, the system
achieved a % accuracy rate for classifying anesthesia depth. used by the anesthesiologists such as blood pressure, la