基於模糊理論之中醫辨證

A Fuzzy Syndrome Differentiation System in Traditional Chinese Medicine

李培寗(Pei-Ning Li)

指導教授 ： 林迺衛

國立中正大學/工學院/資訊工程學系/碩士(2015年)

摘要

中醫辨證體系非常龐雜，一般醫生臨床難以迅速掌握全貌。本研究希望藉由計算機龐大的記憶能力及迅速的分析能力，研製一個中醫自動辨證系統，做為中醫辨證的研究及教學平台。
《中醫證候學》[1]整理古今中醫辨證文獻，提出一個統一的辨證體系。它先依病因為綱將疾病分為15個證門及45個證類。它接著依病位為目將45個證類再細分為281證型。它最後依病機為科將281個證型細分為2344個證候。本論文根據《中醫證候學》[1]的虛證門，研製一個中醫自動辨證系統。
中醫自動辨證系統需要決定每一個證的特徵症狀集合及設計證的分類機制。本論文考量中醫辨證本具的模糊性，運用模糊集合來表示證的特徵症狀集合，並運用模糊分類來設計證的分類機制。本論文也做了一個初步的實驗來評估使用模糊理論及未使用模糊理論的辨證差異。

關鍵字

模糊邏輯 ； 中醫辨證 ； 模糊理論

並列摘要

The syndrome differentiation system in Traditional Chinese Medicine (TCM) is huge and complex. It is very difficult for a doctor to apply efficiently and completely such a syndrome differentiation system in clinical diagnosis. This research aims to utilize the large amount of storage and the efficient computing power of computers to develop an automatic syndrome differentiation system. We wish to use this automatic syndrome differentiation system as a platform for TCM research and teaching.
The book Syndromatology of Traditional Chinese Medicine collects and reorganizes many ancient and modern literatures on the theory and practice of syndrome differentiation in traditional Chinese medicine. This book proposes a uniform syndrome differentiation system. This system categorizes syndromes into 15 syndrome divisions and 45 syndrome subdivisions according to the causes of diseases. These 45 syndrome subdivisions are then divided into 281 syndrome classes according to locations of diseases. Finally, these 281 syndrome classes are divided into 2344 syndromes according to mechanisms of diseases. This thesis developed an automatic syndrome differentiation system according to the deficiency syndrome division.
The automatic syndrome differentiation system needs to determine the set of symptoms for each syndrome and design a mechanism for syndrome differentiation. This thesis recognizes the fuzziness in the syndrome differentiation. It uses fuzzy sets to represent the set of symptoms for each syndrome and uses fuzzy classification to implement the mechanism for syndrome differentiation. This thesis also provides the result of a preliminary experiment on evaluating the differences on syndrome differentiation between the approach using fuzzy theory and the one without using fuzzy theory.

並列關鍵字

Tradictional Chinese Medicine ； syndrome differentiation ； fuzzy logic

參考文獻

[6] 陳南宏，”中醫症狀知識本體的建構”，碩士論文，國立中正大學資訊工程學系，台灣，2015

[7] 游淑晴，”中醫虛證辨證系統”，碩士論文，國立中正大學資訊工程學系，台灣，2014

[8] 蕭志豪，”中醫辨證系統”，碩士論文，國立中正大學資訊工程學系，台灣，2014

[9] 劉建宏，”奠基於貝氏網路的中醫證型辨識”，碩士論文，國立中正大學資訊工程學系，台灣，2012

[12] J. Gong, S. Lu, R. Wang, “Pdhms: Pulse diagnosis via wearable healthcare sensor networks,” Proceedings of IEEE International Conference on Communications, June 2011, pp.1-5.