Year: 2012
Type: article
Abstract: This paper presents a method for automatic segmentation of the prostate from transversal T2-weighted images based on 3D Active Appearance Models (AAM). The algorithm consist of two stages. Firstly, Shape Context based non-rigid surface registration of the manual segmented images is used to obtain the point correspondence between the given training cases. Subsequently, an AAM is used to segment the prostate on 50 training cases. The method is evaluated using a 5-fold cross validation over 5 repetitions. The mean Dice similarity coefficient and 95% Hausdorff distance are 0.78 and 7.32 mm respectively.
Authors: Bianca Maan, Ferdinand van der Heijden
Language: English
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W1574802394