Title :
Deduction research on syndromes diagnosis of TCM Inquiry for Cardiovascular Diseases based on RBF nerve network
Author :
Xu, Zhao-Xia ; Wang, Yi-Qin ; Liu, Guo-Ping ; Xu, Jin ; Guo, Rui ; Yan, Hai-Xia ; Li, Fu-Feng ; Hong, Yu-jian ; Yan, Jian-jun ; Xia, Chun-ming
Author_Institution :
Shanghai Univ. of Traditional Chinese Med., Shanghai, China
Abstract :
In this study, we collected cases by means of uniform Information- Collection Scale of Inquiry Diagnosis for Cardiovascular Diseases in TCM. The diagnostic criteria were established, and each case was interpreted by associate chief physician or senior doctor. The value “1 or 0” was assigned to “with or without” information. By using Epidate 3.1, the information collected was entered by two persons and twice each, and the database was established. Dialectical deduction research of inquiry diagnosis for cardiovascular diseases on the basis of RBF nerve network was conducted. The RBF nerve network parameters were determined after many experiments: (1) velocity was between 0.1-5; (2) target error was 0.00001. The result showed that the overall precision rate was as high as 88.86% for identification of syndromes of cardiovascular diseases in TCM on the basis of RBF nerve network. The precision rate for RBF nerve network to recognize syndromes of TCM was rather high, which could provide theoretical and technical support for the objectivity and standardization study of syndromes of TCM.
Keywords :
blood vessels; cardiology; diseases; medical computing; neural nets; patient diagnosis; radial basis function networks; Epidate 3.1; RBF nerve network; TCM; cardiovascular diseases; diagnostic criteria; dialectical deduction; inquiry diagnosis information collection scale; neural network parameters; syndrome diagnosis; traditional Chinese medicine; Cardiovascular Diseases; RBF Nerve Network; Syndromes Diagnosis; TCM Inquiry;
Conference_Titel :
Bioinformatics and Biomedicine Workshops (BIBMW), 2010 IEEE International Conference on
Conference_Location :
Hong, Kong
Print_ISBN :
978-1-4244-8303-7
Electronic_ISBN :
978-1-4244-8304-4
DOI :
10.1109/BIBMW.2010.5703887