Title: Reconstruction of a 3D mesh with displacement vectors for medical images

Authors: Said Benomar El Mdeghri; Malki Mohamed Ouçamah Cherkaoui; Abdelhak Kaddari; Abdelali Elloub

Addresses: Department of Computer Science, Faculty of Science Dhar El Mahraz, University Sidi Mohammed Ben Abdellah, Fez, Morocco ' Department of Computer Science, Faculty of Science Dhar El Mahraz, University Sidi Mohammed Ben Abdellah, Fez, Morocco ' Department of Computer Science, Faculty of Science Dhar El Mahraz, University Sidi Mohammed Ben Abdellah, Fez, Morocco ' Department of Computer Science, Faculty of Science Dhar El Mahraz, University Sidi Mohammed Ben Abdellah, Fez, Morocco

Abstract: 2D medical images resolution from the current hardware becomes more important. The reconstruction of a 3D model from these images generates a large number of polygons, which leads to problems with the amount of information stored and also in terms of the complexity of calculations for display viewing in real-time. Our approach to 3D reconstruction of the human anatomy, from a sequence of 2D images is based on a low mesh resolution. This mesh is obtained by using the marching cubes algorithm of these medical images. Small reliefs lost in the base mesh are calculated from the information we have in 2D images. These reliefs are shown in the form of 2D vectors, which allows a very important optimisation of the 3D model storage. Details of each polygon mesh base are calculated in an independent manner, which facilitator the automatic generation of the displacement map for this mesh. The use of this card with the latest GPU accelerates the real-time calculations of visualisation.

Keywords: 3D reconstruction; human anatomy; marching cubes; small reliefs; 2D displacement vectors; storage optimisation; displacement map; real-time visualisation; image reconstruction; 3D mesh; medical images; image processing; 3D modelling.

DOI: 10.1504/IJMEI.2015.070117

International Journal of Medical Engineering and Informatics, 2015 Vol.7 No.3, pp.209 - 221

Received: 11 Apr 2014
Accepted: 08 Aug 2014
Published online: 27 Jun 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article