A wavelet based method for steady-state detection in anesthesia

With the recent concern on patient´s outcome following general anesthesia, automatic control of drugs has been a field of interest. The Bispectral Index (BIS) is an EEG based hypnosis monitor, in current use at the operating theatre as a guiding tool for the anesthesiologist to titrate drugs, and prevent awareness. When trying to model a certain process, it´s very important to obtain information of the system behavior under steady-state conditions. In this study the hypnotic and analgesic drugs´ effect on the hypnosis index BIS was analyzed, in order to obtain steady-state information of the system response (inputs-output), and in the future model the drugs combined effect. A steady-state index was obtained using a wavelet analysis technique for trend detection. This tool may be used in the future to model the drug´s combined effect on the hypnosis indices, and also to bring some insight on disturbances not related to drug changes.