心臟病是全球十大死因之一。而心臟疾病中，又以冠狀動脈疾病(CAD)為主要的形式。雖然心導管檢查可以得到較精準的診斷結果，但是這種方法卻很昂貴且有可能傷到病人。非侵入式的診斷方法可以降低病人受傷的風險，但有準確度下降，或是耗時而且所費不貲等問題。一個準確、低成本且快速的診斷方式是有需要的。

本論文旨在提供一個使用適應性類神經模糊推論系統(ANFIS)演算法的模型，一個具有利使用練資料來自我修正的能力的模型，以建立可以用來運用在冠心症的診斷系統，一個能夠模擬醫生藉著病人提供之有限的口語化資訊來判斷病情的專家系統。這套系統將分為兩個部分，第一部分為取得輸入資料，檢查輸入資料是否有缺失。如果有的話，系統要將缺失修正。第二部分使用ANFIS演算法來學習如何診斷。

ANFIS模型使用倒傳遞演算法結合最小平方法進行訓練。四種等級的冠狀動脈疾病結果，依照嚴重程度來分類，被診斷系統用來輔助醫生選擇最合適的治療方式。訓練的表現和分類的準確度將被用來評估ANFIS模型的性能。最後和模糊診斷系統的結果進行比較。結果顯示，ANFIS模型的表現 (80.6%)優於模糊診斷系統的表現 (72.7%)。

Heart disease is one of the top ten causes of death in global, and coronary artery disease (CAD) is the main form of heart disease. Cardiac catheterization gives accurate results, but it is expensive and may be harmful to patients. Non-invasive methods can reduce damage risk but have lower accuracy and other problems like time-consuming and expensive. Therefore, a diagnosis method that is accurate, cost-effective, and time-saving is desirable.

In this thesis, a model that uses an adaptive neural fuzzy inference system (ANFIS) is presented, which is able to build a diagnosis system with self-correction by training data. An expert system for heart diseases that follows doctor judgement from limited linguistic information given by patients is applied. There are two phases in the system. In the first phase, input features are obtained, checking if input features have defect. If so, the system has to correct the defect. In second phase, an ANFIS algorithm is used for classification.

The ANFIS model is trained by using the back propagation method which combines with the least squares method. Four levels of CAD results which are classified by the severity of the disease, are used by the diagnosis system to help doctors choose the most appropriate treatment for patients. The training performance and classification accuracies are used to evaluate the performance of the ANFIS model. The performance of ANFIS model is compared with the fuzzy diagnosis system. The results shows that the accuracy of the ANFIS model (80.6%) is better than the accuracy of the fuzzy model (72.7%).

[1] Tsipouras, M.G., et al., Automated diagnosis of coronary artery disease based on data
mining and fuzzy modeling. Information Technology in Biomedicine, IEEE Transactions
on, 2008. 12(4): p. 447-458.
[2] Yan, H., et al., A multilayer perceptron-based medical decision support system for heart
disease diagnosis. Expert Systems with Applications, 2006. 30(2): p. 272-281. 1969.
[3] Douali, N., et al., Diagnosis support system based on clinical guidelines: comparison
between case-based fuzzy cognitive maps and Bayesian networks. Computer methods
and programs in biomedicine, 2014. 113(1): p. 133-143.
[4] Polat, K. and S. Güneş, An expert system approach based on principal component analysis
and adaptive neuro-fuzzy inference system to diagnosis of diabetes disease. Digital
Signal Processing, 2007. 17(4): p. 702-710.
[5] Marateb, H.R. and S. Goudarzi, A noninvasive method for coronary artery diseases
diagnosis using a clinically-interpretable fuzzy rule-based system. Journal of research
in medical sciences: the official journal of Isfahan University of Medical Sciences,
2015. 20(3): p. 214.
[6] Acharya, U.R., et al., Automated diagnosis of glaucoma using texture and higher order
spectra features. Information Technology in Biomedicine, IEEE Transactions on, 2011.
15(3): p. 449-455.
[7] Khatibi, V. and G.A. Montazer, A fuzzy-evidential hybrid inference engine for coronary
heart disease risk assessment. Expert Systems with Applications, 2010. 37(12): p. 8536-
8542.
[8] Hartigan, J.A. and M.A. Wong, Algorithm AS 136: A k-means clustering algorithm.
Journal of the Royal Statistical Society. Series C (Applied Statistics), 1979. 28(1): p.
100-108.
[9] Zadeh, L.A., Toward a theory of fuzzy information granulation and its centrality in
human reasoning and fuzzy logic. Fuzzy sets and systems, 1997. 90(2): p. 111-127.
[10] Mamdani, E.H. and S. Assilian, An experiment in linguistic synthesis with a fuzzy logic
controller. International journal of man-machine studies, 1975. 7(1): p. 1-13.
[11] Takagi, T. and M. Sugeno, Fuzzy identification of systems and its applications to modeling
and control. Systems, Man and Cybernetics, IEEE Transactions on, 1985(1): p.
116-132.
[12] Hansen, L.K. and P. Salamon, Neural network ensembles. IEEE Transactions on Pattern
Analysis and Machine Intelligence, 1990(10): p. 993-1001.
[13] Rumelhart, D.E., J.L. McClelland, and P.R. Group, Parallel Distributed Processing:
Explorations in the Microstructure of Cognition, Vol. 1-2. Cambridge, MA, 1986.
[14] Jang, J.-S.R., ANFIS: adaptive-network-based fuzzy inference system. Systems, Man
and Cybernetics, IEEE Transactions on, 1993. 23(3): p. 665-685.
[15] Höskuldsson, A., PLS regression methods. Journal of chemometrics, 1988. 2(3): p.
211-228.
[16] Jang, J.-S.R. Input selection for ANFIS learning. in Proceedings of the fifth IEEE international
conference on fuzzy systems. 1996. Citeseer.
[17] Stavrakoudis, D.G., I.Z. Gitas, and J.B. Theocharis. A hierarchical genetic fuzzy
rule-based classifier for high-dimensional classification problems. in Fuzzy Systems
(FUZZ), 2011 IEEE International Conference on. 2011. IEEE.
[18] Kohavi, R. and F. Provost, Confusion matrix. Machine learning, 1998. 30(2-3): p. 271-
274.
[19] Bradley, A.P., The use of the area under the ROC curve in the evaluation of machine
learning algorithms. Pattern recognition, 1997. 30(7): p. 1145-1159.
[20] Horikawa, S.-i., T. Furuhashi, and Y. Uchikawa, On fuzzy modeling using fuzzy neural
networks with the back-propagation algorithm. IEEE transactions on Neural Networks,
1992. 3(5): p. 801-806.
[21] Kohavi, R. A study of cross-validation and bootstrap for accuracy estimation and model
selection. in Ijcai. 1995.
[22] Yang, J.-G., et al., Coronary heart disease optimization system on adaptive-networkbased
fuzzy inference system and linear discriminant analysis (ANFIS–LDA). Personal
and Ubiquitous Computing, 2014. 18(6): p. 1351-1362.