This paper presents a new approach to detect depth of anaesthesia by using recurrence quantification analysis of electroencephalogram (EEG) and artificial neural network(ANN) . From 98 consenting patient experiments, 98 distinct EEG recordings were collected prior to incision during isoflurane anaesthesia of different levels. The seven measures of recurrence plot were extracted from each of four-channel EEG time series. Prediction was made by means of ANN. Training and testing the ANN used the ‘leave-one-out’ method. The prediction was tested by monitoring the responses to incision. The system was able to correctly classify purposeful responses in average accuracy of 92.86% of the cases. This method is also computationally fast and acceptable real-time clinical performance was obtained.