- 學位論文
應用基因演算法及模糊邏輯技術來監控麻醉深度
Monitoring and control the depth of anesthesia using genetic algorithms and fuzzy logic
若您是本文的作者，可授權文章由華藝線上圖書館中協助推廣。
摘要
麻醉自動化近年來越來越受到外國的重視，這門學問並不是單單由麻醉醫師或是控制工程師所能夠獨立做到的，這是一項跨領域的技術整合應用，其中最大的難處在於麻醉深度的判別，這也是許多研究學者想在麻醉醫學方面有所突破的重要方向。然而，開刀房中麻醉深度的判別，不單單只是為了知道病人目前現在的麻醉狀況，更是為了進一步來控制病人所需的麻醉藥量，使得病人能在最安全的狀態下完成手術。 本論文便是根據現有的環境和設備，配合台大醫學院麻醉科的麻醉醫師，來完成手術中病人之監視及麻醉深度自動控制的智慧型系統。而首要的工作就是利用智慧型之類神經網路理論、模糊理論、基因演算法及群集分析來建立靜脈麻醉藥Propofol對於人體生理信號的模型，而本論文所篩選的輸入訊號有體重（Weight）和累加過後的Propofol靜脈注射藥量流率（Sum_PIR）等；輸出訊號有血壓（SAP）、心跳（HR）、量測雙頻譜分析指標Bispectral Index（BIS）等，將病人麻醉模型建構好後，再來則是進行模擬控制，確認所採用之比例-積分-微分控制器(Proportional-Integral-Derivative, PID)及Fuzzy控制器能夠穩當的進行麻醉控制。最後再進行本論文的重點臨床試驗，並已完成五個成功案例。
並列摘要
Anesthesia is a taxing discipline for application of automated on-line drug infusion. To keep the safety of the patient during surgical operations, it is important to on-line monitor the anesthetic condition of the patient. However, Depth of anesthesia (i.e. unconsciousness) is harder to define and not readily measurable during surgery. In practice, anesthetists have a number of clinical signs and on-line measurements which can be used selectively for the determination of the patient's state. Therefore, many methods have been used for monitoring of anesthetic depth based on different clinical measurements. In order to simulate the whole operation during intravenous anesthesia, two inputs (i.e., weight and summation of propofol infusion rate) and three outputs (i.e., blood pressure, heart rate and bispectral index signals) fuzzy model with genetic algorithms and clustering have been designed for the patient model. Also, the patient model has been validated and simulated by proportional -integral-derivative (PID) controller and fuzzy logic controller. Furthermore, clinical trials have been tested in five cases during intravenous anesthesia.
參考文獻
1. S. Charbonnier, S. Galicher, G. Mauris, J. P. Siche, “Statistical and Fuzzy Model of Ambulatory Systolic Blood Pressure for Hypertension Diagnosis,” IEEE Transactions on Instrumentation and Measurement, Vol. 49, No. 5, Oct., pp. 998-1003, 2000.
2. N. E. Mansour, D. A. Linkens, “Self-tunning Pole-placement Multivariable Control of Blood Pressure for Post-operative Patients: A Model-based Study,” IEE Proceedings-Control Theory and Applications, Vol. 137, Jan., pp. 13 —29, 1990.
3. J. R. Jauchem, M. R. Frei, K. L. Ryan, J. H. Merrit, M. R. Murphy, “Lack of Effects on Heart Rate and Blood Pressure in Ketamine-Anesthetized Rats Briefly Exposed to Ultra Wideband Electromagnetic Pulses,” IEEE Transactions on Biomedical Engineering, Vol. 46, No. 1, Jan., pp. 117-120, 1999.
4. R. J. Gajrajj, M. Doi, H. Mantzaridis, G. N. C. Kenny, “Analysis of the EEG bispectral, auditory evoked potentials and the EEG Power Spectrum During Repeated Transitions From Consciousness to Unconscoousness,” British Journal of Anaesthesia, 80: pp. 46-52, 1998.
5. P. S. Glass, M. Bloom, L. Kearse, C. Rosow, P. Sebel, P. Manberg, “Bispectral Analysis Measure Sedation and Memory Effects of Propofol, Midazolam, Isoflurance, and Alfentanil in Healthy Volunteers,” Anesthesiology, V 86, No. 4, Apr., pp. 836-847, 1997.