Comparison of 2D and 3D region-based deformable models and random walker methods for PET segmentation

摘要: In this paper, we propose to compare different methods for tumor segmentation in positron emission tomography (PET) images. We first tackle problem under the umbrella of shape optimization and 3D deformable models. Indeed, 2D active contours have been widely investigated literature but these techniques do not take advantage informations. On one hand, use well-known model Chan Vese. other hand a criterion based on parametric probabilities which allows us test assumption Poisson distribution intensity such Both will be compared their equivalent an improved random-walker algorithm. For comparison, set simulated, phantom real sequences with known ground-truth compute corresponding Dice Coefficients. also give some examples results.