Modelling the hypnotic patient response in general anaesthesia using intelligent models

Current trends in modern medicine are concerned with the development of the so-called personalized medicine whose aim is tailoring of treatment to patients. The present research is focused on anaesthesiology, specifically on the study of the level of unconciousness (hypnosis) of patients under general anaesthesia. The idea is to improve the prediction of the patient response during anaesthesia, so that the clinician could adjust the drug dosing using this information. The bispectral index (BIS) is a signal provided by electroencephalogram monitors that is accepted as a depth of hypnosis index. This work shows the BIS modelling of patients undergoing general anaesthesia during surgery. For this, a model that allows to know its value from the electromyogram (EMG) and the propofol infusion rate has been created. The proposal has been achieved by using clustering combined with regression techniques and using a real data set obtained from patients undergoing general anaesthesia. Finally, the created model has been tested also with data from real patients, and the results obtained attested the accuracy of the model.