Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Cited by Google
- Similars in SciELO
- Similars in Google
Share
Acta Biológica Colombiana
Print version ISSN 0120-548X
Abstract
TIBAMOSO, GERARDO; RUEDA, ANDREA and ROMERO, EDUARDO. Semi-Automatic Liver Volume Segmentation in Computed Tomography Images. Acta biol.Colomb. [online]. 2010, vol.15, n.3, pp.261-274. ISSN 0120-548X.
Liver volume is a significant parameter in surgery for tumor extraction, transplants, and regeneration treatments. Generally, the volume estimation is obtained from manual segmentations performed by specialists, resulting in a tedious process with low reproducibility. In this work a semi-automatic method for the liver volume segmentation in CT images is presented. The method consist in manually superimpose a triangular surface on the images, and use a movement equation associated to each vertex to deform the surface and delimit the liver boundaries. Surface dynamics depend on intensity and gradient information, and neighboring relationships between vertices, until a fixed number of iterations is reached. Comparison between the obtained results and reference segmentations in 20 CT scans, show the surface adaptability to the shape and the diffuse boundaries of the liver, two of the principal segmentation problems.
Keywords : Liver segmentation; deformable surfaces.