Enhanced Algorithm for Reconstruction of Three-Dimensional Mesh from Medical Images using Tessellation of Recent Graphics Cards

The reconstruction of a 3D mesh using displacement vectors for medical images is a recent method that allows the exploitation of modern GPUs. This method demonstrated its efficiency by accelerating 3D visualization calculations and optimizing the storage process. In fact, it is divided into two main stages. The first step is the construction of a basic mesh by applying the Marching Cubes algorithm, and the second step is the extraction of the displacement vectors, which represent the details lost in the basic mesh. In fact, the Marching Cubes algorithm used to build the basic mesh suffers from some problems that we will try to overcome in this article. These problems are summarized in the ambiguity encountered during the construction of the basic mesh in some cases. Also, the resulting basic mesh must undergo modifications, in order not to have errors of form, which requires time and memory, and which gives the end a final mesh which is not optimal and even erroneous in certain situations. Our method is based on extracting the contours of the anatomy to be reconstructed from a sequence of 2D images. Each contour will be represented by a triangle. The shape of the basic mesh will then be the result of the connection of these triangles. This strategy avoids the use of the marching cubes algorithm in the reconstruction of the basic mesh in order to overcome the problems mentioned above.

@article{Miara2020,
  title     = {Enhanced Algorithm for Reconstruction of Three-Dimensional Mesh from Medical Images using Tessellation of Recent Graphics Cards},
  journal   = {International Journal of Advanced Computer Science and Applications},
  volume    = {11},
  number    = {12},
  year      = {2020},
  publisher = {The Science and Information Organization},
  author    = {Lamyae Miara and Said Benomar El Mdeghri and Mohammed Oucamah Cherkaoui Malki},
  doi       = {10.14569/IJACSA.2020.0111263},
  url       = {https://doi.org/10.14569/IJACSA.2020.0111263}
}