The undersigned hereby assign all rights, included but not limited to copyright, for this manuscript to CMB Association upon its submission for consideration to publication on Cellular and Molecular Biology. The rights assigned include, but are not limited to, the sole and exclusive rights to license, sell, subsequently assign, derive, distribute, display and reproduce this manuscript, in whole or in part, in any format, electronic or otherwise, including those in existence at the time this agreement was signed. The authors hereby warrant that they have not granted or assigned, and shall not grant or assign, the aforementioned rights to any other person, firm, organization, or other entity. All rights are automatically restored to authors if this manuscript is not accepted for publication.
Parameter estimation and order selection for an empirical model of VO2 on-kinetics.
Corresponding Author(s) : O. Alata
Cellular and Molecular Biology,
Vol. 53 No. 2: Biomedical signal and image processing - Volume 2
Abstract
In humans, on-kinetics are noisy numerical signals that reflect the pulmonary oxygen exchange kinetics at the onset of exercise. They are empirically modelled as a sum of an offset and delayed exponentials. The number of delayed exponentials; i.e. the order of the model, is commonly supposed to be 1 for low-intensity exercises and 2 for high-intensity exercises. As no ground truth has ever been provided to validate these postulates, physiologists still need statistical methods to verify their hypothesis about the number of exponentials of the on-kinetics especially in the case of high-intensity exercises. Our objectives are first to develop accurate methods for estimating the parameters of the model at a fixed order, and then, to propose statistical tests for selecting the appropriate order. In this paper, we provide, on simulated Data, performances of Simulated Annealing for estimating model parameters and performances of Information Criteria for selecting the order. These simulated Data are generated with both single-exponential and double-exponential models, and noised by white and Gaussian noise. The performances are given at various Signal to Noise Ratio (SNR). Considering parameter estimation, results show that the confidences of estimated parameters are improved by increasing the SNR of the response to be fitted. Considering model selection, results show that Information Criteria are adapted statistical criteria to select the number of exponentials.
Keywords
VO2 on-kinetics
parametric modelling
parameter estimation
model selection
optimisation
stochastic methods
general linear model
simulated annealing
information criteria.
Alata, O., & Bernard, O. (2007). Parameter estimation and order selection for an empirical model of VO2 on-kinetics. Cellular and Molecular Biology, 53(2), 11–21. Retrieved from http://mail.cellmolbiol.org/index.php/CMB/article/view/1113
Download Citation
Endnote/Zotero/Mendeley (RIS)BibTeX