Monitoring the depth of anaesthesia using detrended fluctuation analysis

Detrended fluctuation analysis was used to study electroencephalogram fluctuations in order to assess the depth of anaesthesia and measure the level of consciousness. Firstly, the fluctuation function F(s) was calculated, we used the distribution of F(s) in segments of lengths to classify anaesthesia state levels into awake, light, moderate and deep states. Secondly, a linear fit on F(s) versus s in each segment was performed. Finally, the fitted line crossed the defined line to divide four zones corresponding with four anaesthesia states from 100 to 0 in the coordinate system. Experimental results demonstrated that it could identify different anaesthesia states based on electroencephalogram signals accurately, and these state values changed with the clinical situation rapidly. It could also extend the ranges of anaesthesia state values through adjustable parameters, which improved the adaptability of algorithm. The method proposed in this paper provides an effective application in monitoring the depth of anaesthesia.