Patient-customized 3D reconstruction of human ribs from lung MDCT dataset

Author

Pazokifard, Banafsheh ; Sowmya, Arcot ; Moses, Daniel

Author_Institution

Sch. of Comput. Sci. & Eng., Univ. of New South Wales, Sydney, NSW, Australia

fYear

2014

fDate

April 29 2014-May 2 2014

Firstpage

1131

Lastpage

1135

Abstract

An accurate algorithm for 3D rib reconstruction of an individual patient is presented. Segmented ribs are reconstructed in a cylindrical coordinate system, based on available knowledge of human rib geometry. Each rib is reconstructed individually using data from multidetector computed tomography (MDCT) images, which provides geometrical data including unique angle, length, curvature and mass. The proposed algorithm was tested on the datasets from 15 patients. Qualitative and quantitative evaluations show high similarity between reconstruction results and reference masks provided by an experienced radiologist.

Keywords

computerised tomography; image reconstruction; image segmentation; lung; medical image processing; curvature; cylindrical coordinate system; geometrical data; human rib geometry; length; lung MDCT dataset; mass; multidetector computed tomography; patient-customized 3D rib reconstruction; reference masks; segmented ribs; unique angle; Geometry; Image reconstruction; Image segmentation; Ribs; Shape; Smoothing methods; Three-dimensional displays; 3D; MDCT; Reconstruction; anatomically-faithful; automatic; geometry; patient-customized; rib;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Imaging (ISBI), 2014 IEEE 11th International Symposium on

Conference_Location

Beijing

Type

conf

DOI

10.1109/ISBI.2014.6868074

Filename

6868074