Segmentation et recalage d'images en imagerie cardiaque

Autor:	R.H. Rakotobe, C. Fernandez-Maloigne, F. Langevin, M. Fauchet
Rok vydání:	1997
Předmět:	Surface (mathematics) Active contour model genetic structures Computer science business.industry Image processing Applied Microbiology and Biotechnology Edge detection Visualization A priori and a posteriori Computer vision Segmentation Artificial intelligence business Cardiac imaging Biotechnology
Zdroj:	RBM-News. 19:90-93
ISSN:	0222-0776
DOI:	10.1016/s0222-0776(97)81938-4
Popis:	The presented paper concern echocardiography and myocardial SPECT images segmentation and fusion. The left ventricle surface is extracted from SPECT data using tools of differential geometry and active contour models. Perfusion mapping on this surface allow a 3D visualization of cardiac SPECT studies. A deformable template is used for the segmentation of echography images. We propose also a modelisation of left ventricle contours with a bounded algebraic curve. Multimodality images registration is obtained by fitting the (echography) contour to the (SPECT) surface. A first approach based on a priori knowledge is used. The result is refined thanks to a contour to surface registration algorithm. Clinical applications concerm myocardial viability study. The three-dimensional SPECT studies visualization is expected to simplify clinical interpretation.
Databáze:	OpenAIRE
Externí odkaz:	https://explore.openaire.eu/search/publication?articleId=doi_________::a5e65918e707033755a125c539574974 https://doi.org/10.1016/s0222-0776(97)81938-4 Zobrazit plný text záznamu Full Text from ScienceDirect