Segmentation of the Human Trachea Using Deformable Statistical Models of Tubular Shapes

Abstract. In this work, we present two active shape models for the segmentation of tubular objects. The first model is built using cylindrical parameterization and minimum description length to achieve correct correspondences. The other model is a multidimensional point distribution model built from the centre line and related information of the training shapes. The models are used to segment the human trachea in low-dose CT scans of the thorax and are compared in terms of compactness of representation and segmentation effectiveness and efficiency. Leave-one-out tests were carried out on real CT data.

• Type: Conference
• Year: 2007
• Where: ACIVS
• Authors: Romulo Pinho, Jan Sijbers, Toon Huysmans
• Keywords: ACIVS 2007, Active Shape Models, Computer Vision, Low-dose Ct Scans, first Model