BACKGROUND: Bispectrum and bicoherence analysis that is the frequency domain methods of EEG analysis was examined to elucidate the correlation with the depth of anesthesia.

METHODS: 25 patients under the gynecological surgery were studied. EEG was surveyed by PhysioLab 400. Analysis of EEG was done according to the progress of anesthesia, normal state before anesthesia, before incision of skin, operation 1, operation 2, awake and post-anesthesia. EEG is applied to filtering of signal, base line correction, linear detrend removal to get more reliable analysis. The appearance rate of bispectrum peaks are calculated by bispectrum operation. And bicoherences are calculated by bispectrum and power spectrum of AR model.

RESULTS: In awaking state of patient in pre-anesthesia and post anesthesia, appearance rate of bispectral coupling was observed that a strong appearance rate was represented in all area (0-35 Hz). During the anesthesia, the low frequency area (3-15 Hz) revealed a strong appearance. Also the 2D map of a appearance rate is divided into four sections for more objectivity. Of four sections Bipara♯4 is considered that the best parameter which showed progress of anesthesia (P < 0.01). In bicoherence study, in awake state which patient is in pre-anesthesia and post anesthesia, appearance rate of bicoherence value was observed that a strong appearance rate was represented in high frequency range (15-30 Hz). However, under the anesthesia of patients, the low frequency area (0-10 Hz) revealed a strong appearance. And, of four sections Bicpara♯2 and Bicpara♯4 are considered that the best parameter which showed progress of anesthesia effectively (P < 0.01).

CONCLUSIONS: The analysis of the peak appearance of bispectrum and bicoherence is useful monitoring for the depth of anesthesia.