Segmentation of 3D deformable objects with level set based prior models

We propose a level set based deformable model for the segmentation of multiple objects from 3D medical images using shape prior constraints. As an extension to the level set distribution model of object shape presented in [1][2][3], this paper evaluates the performance of the level set representation of the object shape by comparing it with the point distribution model(PDM)[4] using the Chi-square test. We define a Maximum A Posteriori(MAP) estimation model using level set based prior information to realize the segmentation of the multiple objects. To achieve this, only one level set function is employed as the representation of the multiple objects of interest within the image. We then define the probability distribution over the variations of objects contained in a set of training images. We found the algorithm to be computationally efficent, robust to noise, able to handle multidimensional data, and avoids the need for explicit point correspondences during the training phase. Results and validation from various experiments on 2D/3D medical images are demonstrated.