The present invention provides a method of generating index for determining anesthesia consciousness awakening level using artificial neural network, which includes the following steps: first, measuring and obtaining the physiological signal (brain wave signal or eye movement signal) of a subject during a complete surgery period next, filtering the noise of the obtained physiological signal using the empirical mode decomposition (EMD) method segmental calculation of the sample entropy of the noise-filtered physiological signal to obtain a set of sample entropy of physiological signal for the subject then, based on the aforementioned steps performing physiological signal measurement, filtering and calculation of sample entropy to a plurality of subjects to obtain the sets of sample entropy of physiological signal for the plurality of subjects and, performing the artificial neural networks (ANNs) regression analysis using the sample entropy set and the consciousness awakening level obtained by the anesthesia depth monitor during surgery to generate the anesthesia depth index model for determining the consciousness awakening level of a patient after anesthetization in the operating room.本發明提供一種使用類神經網路判斷麻醉意識清醒程度的方法,受測者於一完整手術期間之生理訊號(腦波訊號或眼動訊號)首先被量測取得,接著,將取得之生理訊號經過經驗模態分解(Empirical Mode Decomposition,EMD)法濾除雜訊後,再將濾除雜訊後的生理訊號,分段進行樣本熵值計算後,得到一受測者的生理訊號樣本熵值集合,然後,依據上述步驟進行複數個受測者的生理訊號量測、濾波、樣本熵值計算後,得到複數個受測者的生理訊號樣本熵值集合,將該樣本熵值集合與手術期間經由麻醉深度監測儀取得之意識清醒度進行類神經網路(artificial neural networks,ANNs)迴歸分析後,產生麻醉深度指標模型,以應用於開刀房中判斷病患麻醉後之意識清醒程度。
