A Study On The Depth Of Anesthesia Monitoring Based On Fuzzy Neural Network

During surgery, adequate DOA is very important for patients. Over anesthesia may bring patients with nerve sequela and even death. While insufficient anesthesia may cause Intraoperative awareness, which can often lead to untoward psychological consequences. A monitor capable of estimating DOA is, therefore, desirable for assisting the anesthesiologists in minimizing such incidence. However, the significance of the traditional signs of DOA commonly used by the anesthesiologists such as blood pressure, lacrimation, facial grimacing, movement and so on , and those clinical signs have largely depended on the skeletal muscle activity. With the introduction of the concept of balanced anesthesia using multiple drugs and muscle relaxants, Monitoring DOA represents one of the most controversial and subjective aspect of modern anesthesia, no single indicator has been found to estimate the DOA for all patients and all anesthetic agents. So reliable and noninvasive monitoring of the DOA is highly desirable.So the main focus of this dissertation is on setting up a FNN to fusion the signs of anesthesia, the index of DOA is hoped to built and realize to monitor the DOA. In this dissertation EEG data from patients undergoing general anesthesia with different anesthetic agents, and the group consisted of 31 adults with ages ranging from 25 to 86. Kc, ApEn and WE were extracted from EEG, these parameters were used as an input to the FNN with one output—DOA. The FNN was successfully trained to monitor anesthesia or awake, and the accuracy of output reached 98.66%/93. 86% (anesthesia/normal), the index of DOA was tentatively built, and tentatively realized to monitor DOA.It was proved the FNN was accurate and robust, was a promisingcandidate as an effective tool for monitoring of the DOA and would be help to clinical application. Finally, further research interests in the future were prospected through the results of the research.