DocumentCode :
3493882
Title :
Bridge the gap between syndrome in Traditional Chinese Medicine and proteome in western medicine by unsupervised pattern discovery algorithm
Author :
Wang, Wei ; Zhao, Huihui ; Chen, Jianxin ; Chen, Jing ; Xi, Guangcheng
Author_Institution :
Beijing Univ. of Chinese Med., Beijing
fYear :
2008
fDate :
6-8 April 2008
Firstpage :
745
Lastpage :
750
Abstract :
Studying the molecular basis of syndrome in traditional Chinese medicine (TCM) is a research hotspot and a challenge for medicine society. In this paper, we combine clinical epidemiology, proteome technique and data mining research to investigate the molecular basis of syndrome. We do a clinical epidemiology survey of coronary heart disease to collect case patients and control patients. We also analysis the two-dimensional electrophoresis results of blood samples of included patients to find out the proteins with significant expression. We find out that the blood stasis syndrome has significant association with 10 inflammatory factors proteins. Based on the collected data, we proposed an unsupervised pattern discovery algorithm to detect the significantly associated patterns in the data. 14 patterns containing syndrome and proteins are retrieved, which can be considered as the evidence of association between syndrome of TCM and proteome. Furthermore, we validate the unsupervised pattern discovery results by combining support vector machine and 10-fold cross validation, finding that the accuracy of classifying is higher than 90%, which indicates that the pattern discovery results is believable. The research effort here presents a better insight to the integration of TCM and western medicine and develops a better way to study the molecular basis of syndrome.
Keywords :
cardiology; data mining; diseases; medical computing; molecular biophysics; proteins; support vector machines; unsupervised learning; blood stasis syndrome; clinical epidemiology survey; coronary heart disease; data mining; inflammatory factor; proteome technique; support vector machine; syndrome molecular basis; traditional Chinese medicine; two-dimensional electrophoresis; unsupervised pattern discovery algorithm; western medicine; Animals; Blood; Bridges; Cardiac disease; Data mining; Medical diagnostic imaging; Mutual information; Proteins; Supervised learning; Support vector machines; Data mining; Pattern discovery; Proteome; Support Vector Machine; Traditional Chinese Medicine;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Networking, Sensing and Control, 2008. ICNSC 2008. IEEE International Conference on
Conference_Location :
Sanya
Print_ISBN :
978-1-4244-1685-1
Electronic_ISBN :
978-1-4244-1686-8
Type :
conf
DOI :
10.1109/ICNSC.2008.4525315
Filename :
4525315
Link To Document :
بازگشت