Catalogo Articoli (Spogli Riviste)

OPAC HELP

Titolo:
A NEW METHOD TO ESTIMATE PARAMETERS OF LINEAR COMPARTMENTAL-MODELS USING ARTIFICIAL NEURAL NETWORKS
Autore:
GAMBHIR SS; KEPPENNE CL; BANERJEE PK; PHELPS ME;
Indirizzi:
UCLA,SCH MED,CRUMP INST BIOL IMAGING,DEPT MOL & MED PHARMACOL,A-222B CIBI,700 WESTWOOD PLAZA LOS ANGELES CA 90095 UCLA,SCH MED,DEPT BIOMATH LOS ANGELES CA 90095 CALTECH,JET PROP LAB,NASA PASADENA CA 91125
Titolo Testata:
Physics in medicine and biology
fascicolo: 6, volume: 43, anno: 1998,
pagine: 1659 - 1678
SICI:
0031-9155(1998)43:6<1659:ANMTEP>2.0.ZU;2-3
Fonte:
ISI
Lingua:
ENG
Soggetto:
BRAIN TRANSFER CONSTANTS; CEREBRAL BLOOD-FLOW; TIME UPTAKE DATA; GRAPHICAL EVALUATION; EMISSION TOMOGRAPHY; N-13 AMMONIA; QUANTIFICATION; CALIBRATION; REGRESSION; PREDICTION;
Tipo documento:
Article
Natura:
Periodico
Settore Disciplinare:
Science Citation Index Expanded
Citazioni:
31
Recensione:
Indirizzi per estratti:
Citazione:
S.S. Gambhir et al., "A NEW METHOD TO ESTIMATE PARAMETERS OF LINEAR COMPARTMENTAL-MODELS USING ARTIFICIAL NEURAL NETWORKS", Physics in medicine and biology, 43(6), 1998, pp. 1659-1678

Abstract

At present, the preferred tool for parameter estimation in compartmental analysis is an iterative procedure; weighted nonlinear regression. For a large number of applications, observed data can be fitted to sums of exponentials whose parameters are directly related to the rate constants/coefficients of the compartmental models. Since weighted nonlinear regression often has to be repeated for many different data sets, the process of fitting data from compartmental systems can be very time consuming. Furthermore the minimization routine often converges toa local las opposed to global) minimum. In this paper, we examine thepossibility of using artificial neural networks instead of weighted nonlinear regression in order to estimate model parameters. We train simple feedforward neural networks to produce as outputs the parameter values of a given model when kinetic data are fed to the networks' input layer. The artificial neural networks produce unbiased estimates andare orders of magnitude faster than regression algorithms. At noise levels typical of many real applications, the neural networks are foundto produce lower variance estimates than weighted nonlinear regression in the estimation of parameters from mono-and biexponential models. These results are primarily due to the inability of weighted nonlinearregression to converge. These results establish that artificial neural networks are powerful tools for estimating parameters for simple compartmental models.

ASDD Area Sistemi Dipartimentali e Documentali, Università di Bologna, Catalogo delle riviste ed altri periodici
Documento generato il 28/01/20 alle ore 16:01:00