Surface-surface tissue reconstruction and visualization for the magnetic resonance imaging or tomography imaging engineering design

Academic Article


  • This paper presents a surface-surface tissue reconstruction and visualization algorithms of the magnetic resonance or tomography imaging (MRI/CT) for the practical applications to assist in engineering design and diagnostic diseases. The proposed algorithm starts by feeding MRI/CT images slice by slice, applying the region growing segmentation processes and 3D surface reconstruction. An automatic region growing (ARG) process has been proposed in the new framework, and the core of ARG contains a local search algorithm and homogeneity criterions. The initial seeds are obtained by using the local search algorithm while the homogeneity criterion values are selected as a probability function of image pixels. The ARG is used to segment an image into multiple surfaces (2D imaging planes). The surface-surface method is provided to create volume imaging based on marching cube algorithm which is used to obtain vertices and triangulation from segmentation tissue. Due to the diversity of algorithms at reconstruction processes, the resulting surfaces of marching cube need to be smooth and accurate. Therefore, a mean filtering smoothing technique has been applied to smooth the surfaces of tissues for accurate 3D tissue reconstruction. The visualization process is applied to these vertices and triangulations in order to get more accurate 3D reconstruction view of the MRI/CT images. The proposed algorithms are applied to MRI/CT images to prove their efficiency and the assessment of the obtained results is judged visually.
  • Authors

    Digital Object Identifier (doi)

    Author List

  • Altameem T; Zanaty EA; Tolba A; Asaad A
  • Start Page

  • 1462
  • End Page

  • 1468
  • Volume

  • 6
  • Issue

  • 6