Title :
Hormone secretion estimation using fuzzy deconvolution
Author_Institution :
Dept. of Math., Limburg Univ., Maastricht, Netherlands
Abstract :
In this paper we describe a fuzzy identification method to find the instantaneous secretion rate of a hormone. The dynamics of the relation between secretion rate and peripheral plasma hormone concentration is first identified by using learning signals. The fuzzy identification method is based on fuzzy clustering and optimal output predefuzzification. The proposed method leads to a fuzzy inference system which is able to perform the deconvolution, i.e. reconstruction of the unknown secretion rate. The results compare well with other deconvolution methods
Keywords :
deconvolution; fuzzy systems; identification; inference mechanisms; learning (artificial intelligence); medical signal processing; time series; dynamics; fuzzy clustering; fuzzy deconvolution; fuzzy identification; fuzzy inference system; hormone secretion estimation; learning signals; optimal output predefuzzification; plasma hormone concentration; time series; Biochemistry; Deconvolution; Endocrine system; Fluids and secretions; Fuzzy systems; Neodymium; Plasma measurements; Signal processing; Takagi-Sugeno model; Training data;
Conference_Titel :
Fuzzy Systems, 1996., Proceedings of the Fifth IEEE International Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-3645-3
DOI :
10.1109/FUZZY.1996.552720
