ABSTRACT
In this paper 5 methods for the assessment of signal entropy are compared in their capability to follow the changes in the EEG signal during transition from continuous EEG to burst suppression in deep anesthesia. To study the sensitivity of the measures to phase information in the signal, phase randomization as well as amplitude adjusted surrogates are also analyzed. We show that the selection of algorithm parameters and the use of normalization are important issues in interpretation and comparison of the results. We also show that permutation entropy is the most sensitive to phase information among the studied measures and that the EEG signal during high amplitude delta activity in deep anesthesia is of highly nonlinear nature.
Subject(s)
Anesthesia/methods , Electroencephalography/methods , Adult , Algorithms , Anesthesiology/methods , Anesthetics, Inhalation/therapeutic use , Entropy , Fourier Analysis , Humans , Male , Models, Statistical , Monitoring, Physiologic/methods , Normal Distribution , Propofol/pharmacologyABSTRACT
OBJECTIVE: It was hypothesized that somato- sensory evoked potentials can be achieved faster by selective averaging during periods of low spontaneous electroen- cephalographic (EEG) activity. We analyzed the components of EEG that decrease the signal-to-noise ratio of somatosensory evoked potential (SEP) recordings during propofol anesthesia. METHODS: Patient EEGs were recorded with a high sampling frequency during deep anesthesia, when EEGs were in burst suppression. EEGs were segmented visually into bursts, spindles, suppressions, and artifacts. Tibial somatosensory evoked potentials (tSEPs) were averaged offline separately for burst, suppression, and spindle segments using a signal bandwidth of 30-200 Hz. Averages achieved with 2, 4, 8, 16, 64, 128, and 256 responses were compared both visually, and by calculating the signal-to-noise ratios. RESULTS: During bursts and spindles, the noise levels were similar and significantly higher than during suppressions. Four to eight times more responses had to be averaged during bursts and spindles than during suppressions in order to achieve a similar response quality. Averaging selectively during suppressions can therefore yield reliable tSEPs in approximately one-fifth of the time required during bursts. CONCLUSION: The major source of EEG noise in tSEP recordings is the mixed frequency activity of the slow waves of bursts that occur during propofol anesthesia. Spindles also have frequency components that increase noise levels, but these are less important, as the number of spindles is fewer. The fastest way to obtain reliable tSEPs is by averaging selectively during suppressions.
Subject(s)
Anesthesia , Electroencephalography/methods , Adolescent , Adult , Anesthetics, Intravenous , Electromyography/methods , Evoked Potentials, Somatosensory/drug effects , Evoked Potentials, Somatosensory/physiology , Female , Humans , Male , Middle Aged , Monitoring, Intraoperative , Monitoring, Physiologic , Propofol/pharmacology , Tibia/innervationABSTRACT
Electroencephalogram spindle patterns corresponding to two different phenomena-natural sleep and propofol anesthesia-are compared. The spindles are extracted from 5 overnight sleep recordings and 10 recordings of deep propofol anesthesia. Mean frequency, angle of the trend in instant frequency as well as 3 nonlinear parameters-spectral entropy, approximate entropy, and Higuchi fractal dimension- are calculated to characterize the spindle waveforms. Using the Wilcoxon rank sum test with significance level of 0.01, all the mentioned features, except approximate entropy, differ significantly for the two types of EEG spindles.
