@InProceedings{OlivaIsoaMato:2008:BaEsHy,
author = "Oliva, Dami{\'a}n Ernesto and Isoardi, Roberto Andr{\'e}s and
Mato, Germ{\'a}n",
affiliation = "Universidad Nacional de Buenos Aires, Argentina and Escuela de
Medicina Nuclear, Mendoza, Argentina and Grupo F{\'{\i}}sica
Estad{\'{\i}}stica, Centro At{\'o}mico Bariloche, Argentina",
title = "Bayesian estimation of Hyperparameters in MRI through the Maximum
Evidence Method",
booktitle = "Proceedings...",
year = "2008",
editor = "Jung, Cl{\'a}udio Rosito and Walter, Marcelo",
organization = "Brazilian Symposium on Computer Graphics and Image Processing, 21.
(SIBGRAPI)",
publisher = "IEEE Computer Society",
address = "Los Alamitos",
keywords = "Image segmentation, Bayesian analysis, MRI.",
abstract = "Bayesian inference methods are commonly applied to the
classification of brain Magnetic Resonance images (MRI). We use
the Maximum Evidence (ME) approach to estimate the most probable
parameters and hyperparameters for models that take into account
discrete classes (DM) and models accounting for the partial volume
effect (PVM). An approximate algorithm was developed for model
optimization, since the exact image inference calculation is
computationally expensive. The method was validated using
simulated images and a digital phantom. We show that the Evidence
is a very useful figure for error prediction, which is to be
maximized respect to the hyperparameters. Additionally, it
provides a tool to determine the most probable model given
measured data.",
conference-location = "Campo Grande, MS, Brazil",
conference-year = "12-15 Oct. 2008",
doi = "10.1109/SIBGRAPI.2008.5",
url = "http://dx.doi.org/10.1109/SIBGRAPI.2008.5",
language = "en",
ibi = "6qtX3pFwXQZG2LgkFdY/UQ4Vi",
url = "http://urlib.net/ibi/6qtX3pFwXQZG2LgkFdY/UQ4Vi",
targetfile = "Oliva-Bayesian.pdf",
urlaccessdate = "2024, May 02"
}