中西醫診斷專家系統於冠狀動脈心臟病之應用

An Expert System of Coronary Artery Disease in Chinese and Western Medicine

劉德笙(Te-Sheng Liu)

指導教授 ： 翁清松

中原大學/工學院/醫學工程研究所/碩士(2004年)

https://doi.org/10.6840/cycu200400407

摘要

根據行政院衛生署所發佈的「92年國人主要死因與主要癌症統計結果」資料顯示，台灣地區民眾平均每一百個人死亡中就有將近九個人是與心臟疾病有關，長期高居台灣地區主要死亡原因第三位，而心臟疾病之中，又以冠狀動脈心臟病(以下簡稱冠心病)，因為可能引起心肌梗塞或是心因性猝死，對健康的危害最大。因此，冠狀動脈心臟病的診斷與預防是值得來研究的。
本研究主要目的在結合中醫與西醫兩者對於冠心病的臨床醫療診斷方式，給予冠心病症狀定量分型與規範，並藉由人工智慧技術將所隱含的知識擷取出來，以發展適合解決問題的電腦化之專家系統。此外，也利用統計分析的方式來探討中醫證型與西醫生理參數之間的關係，以期望能作為未來冠心病中西醫結合研究中重要的參考因素。
在專家系統的設計上，主要是利用建立起的資料分析系統輔以使用者界面來建立一電腦化中西醫整合系統，並將建立資料庫來儲存相關臨床資料以便後續相關研究之使用。其中在資料分析系統設計方面，主要核心是利用類神經網路、規則萃取與決策串列等技術所架構出來。在類神經網路的部份，採用倒傳遞(Back-Propagation)類神經網路的學習法則來建立，網路共分輸入層、隱藏層與輸出層三層。輸入層為中醫症狀與西醫診斷因子，輸出層分別為中醫證型以及處方用藥。在規則萃取部份，採用教導型(decompositional)演算法。最後，再以決策串列形式的診斷順序輸出系統推論結果。
本研究共收集47名冠心病患者來作為本次研究的對象，採用的症狀共六十九種，證型共有九組，處方共有十一種。結果將所建立的系統在經過資料測試後，得知系統的最佳診斷準確率為95.6％。此外，根據所獲得之規則可發現某些中醫證型與西醫的診斷因素彼此之間有相關性。然而由於系統在推論處方用藥的部份並無法有良好的分析結果，所以在處方用藥的部份只能僅供醫師參考之用。
最後，期望所建立起的輔助系統，對於冠心病醫療能提供一資訊化的途徑來幫助醫師與其它使用者獲得所需之相關資訊。

關鍵字

規則萃取 ； 類神經網路 ； 冠狀動脈心臟病 ； 專家系統

並列摘要

According to the report of Department of Health, the “main cause of death and main cancer statistics of 2003” said that, 9% of people died of heart disease, and which is also the third in death reason in Taiwan for a long time. In all the reasons of heart disease, coronary artery disease (CAD) is the greatest risk for health. It may cause the myocardial infarction or cardiogenic sudden death. Therefore, the diagnosis and prevention of the CAD are very worth to study.
In this research, artificial intelligence technology was used to acquire the implicit knowledge which was obtained by combining the clinical medical diagnosis of both Chinese and Western medicine in the CAD. We would use the knowledge base to develop the computer expert system. In addition, statistical analysis was applied to study the relation of syndrome of Chinese medicine and physiological parameter. The result could be helpful for Chinese and Western medicine about CAD in the future.
The expert system was built by using analysis system and inference engine and to set up the database of expert system to store relevant clinical data. In the analysis system, it was established by neural network, rule extraction and decision list. In the neural network, it was built by Back-Propagation neural network. There were three layers in this neural network. The input layer was the layer for symptoms and the output layer was the layer for syndromes of Chinese medicine output and drugs output. In the rule extraction, a decompositional algorithm was used. Finally, we could get the diagnosis sequence of the decision list from the expert system.
There were 47 CAD patients, 69 symptoms, 9 syndromes, and 11 prescriptions were included in this study. In the test of diagnosis system, the expert system reached a best diagnostic accuracy rate of 95.6%. According to the diagnosis rules we found that some syndromes of Chinese medicine and diagnosis factors of Western medicine were related.

並列關鍵字

Rule extraction ； Artificial neural network ； Coronary artery disease(CAD) ； Expert system

參考文獻

[24]鄭宇茹，中醫聞診結合問診專家系統於慢性腎衰竭之應用，中原大學醫學工程研究所碩士論文，2001。

[34]羅華強，類神經網路-MATLAB的應用，新竹市，清蔚科技股份有限公司，pp.1-2~5-13，2001。

[43]Jen-Cheng Chen, Jia-Sheng Heh, and Chang-Kai Hsu, “Rule extraction of neural networks through hypercube bounds,” Neurocomputing, 2003 (submitted).

[15] K. Saito and R. Nakano, “Medical diagnostic expert system based on PDP model,” ICNN-87, I, pp.255-262, 1987.

[16]Y. Yoon et al., “A connectionist expert system for dermatology diagnosis,” Expert Systems, pp.23-31, 1990.