Study of prognostic factor based on factor analysis and clustering method - Details

Author：

Liu, Zheng (Liu, Zheng.) | Fang, Liying (Fang, Liying.) | Yu, Mingwei (Yu, Mingwei.) | Wang, Pu (Wang, Pu.)

Indexed by：

EI Scopus

Abstract：

Relevance　exists　in　Traditional　Chinese　Medicine(TCM)　clinical　symptoms.　Their　different　combinations　reflect　different　effects.　Focusing　on　these　characteristics,　an　univariate　analysis　method　based　on　the　factor　analysis　and　clustering(FACUA)　is　proposed.　First,　the　independent　common　factors　extracted　from　the　correlative　multivariable　are　used　to　establish　the　eigenvectors　of　symptoms　for　patients.　Then,　the　symptom　patterns　are　discovered　from　the　gathered　similar　symptoms　combination.　The　method　is　verified　by　the　patients　with　advanced　NSCLC(non-small　cell　lung　cancer)　from　Beijing　Hospital　of　Traditional　Chinese　Medicine.　The　experimental　result　shows　that　the　FACUA　method　can　deal　with　the　TCM　clinical　symptoms　and　analyze　the　relationship　between　the　TCM　clinical　symptoms　and　the　tumor　progression.　The　FACUA　method　can　improve　the　universal　applicability　of　the　univariate　analysis　in　TCM　clinical　symptoms.　©　2011　Springer-Verlag　Berlin　Heidelberg.

Keyword：

Factor analysis Medicine Multivariant analysis

Author Community：

[ 1 ] [Liu, Zheng]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
[ 2 ] [Fang, Liying]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China
[ 3 ] [Yu, Mingwei]Beijing Hospital of Traditional Chinese Medicine, CPUMS, Beijing 100124, China
[ 4 ] [Wang, Pu]College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China