Shape Particle Filtering for Image Segmentation
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
Shape Particle Filtering for Image Segmentation. / de Bruijne, Marleen; Nielsen, Mads.
Medical Image Computing and Computer-Assisted Intervention – MICCAI: 7th International Conference, Saint-Malo, France, September 26-29, 2004. Proceedings, Part I. <Forlag uden navn>, 2004. s. 168-175 (Lecture notes in computer science, Bind 3216/2004).Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Shape Particle Filtering for Image Segmentation
AU - de Bruijne, Marleen
AU - Nielsen, Mads
N1 - Conference code: 7
PY - 2004
Y1 - 2004
N2 - Deformable template models are valuable tools in medical image segmentation. Current methods elegantly incorporate global shape and appearance, but can not cope with localized appearance variations and rely on an assumption of Gaussian gray value distribution. Furthermore, initialization near the optimal solution is required. We propose a maximum likelihood shape inference that is based on pixel classification, so that local and non-linear intensity variations are dealt with naturally, while a global shape model ensures a consistent segmentation. Optimization by stochastic sampling removes the need for accurate initialization. The method is demonstrated on three different medical image segmentation problems: vertebra segmentation in spine radiographs, lung field segmentation in thorax X rays, and delineation of the myocardium of the left ventricle in MRI slices. Accurate results were obtained in all tasks.
AB - Deformable template models are valuable tools in medical image segmentation. Current methods elegantly incorporate global shape and appearance, but can not cope with localized appearance variations and rely on an assumption of Gaussian gray value distribution. Furthermore, initialization near the optimal solution is required. We propose a maximum likelihood shape inference that is based on pixel classification, so that local and non-linear intensity variations are dealt with naturally, while a global shape model ensures a consistent segmentation. Optimization by stochastic sampling removes the need for accurate initialization. The method is demonstrated on three different medical image segmentation problems: vertebra segmentation in spine radiographs, lung field segmentation in thorax X rays, and delineation of the myocardium of the left ventricle in MRI slices. Accurate results were obtained in all tasks.
U2 - 10.1007/b100265
DO - 10.1007/b100265
M3 - Article in proceedings
C2 - 20209048
SN - 978-3-540-22976-6
T3 - Lecture notes in computer science
SP - 168
EP - 175
BT - Medical Image Computing and Computer-Assisted Intervention – MICCAI
PB - <Forlag uden navn>
Y2 - 29 November 2010
ER -
ID: 5035008