Comprehensive approach to pulmonary image analysis

Author

Sonka, Milan

Author_Institution

Coll. of Eng., Iowa Univ., Iowa City, IA, USA

fYear

2004

fDate

24-25 June 2004

Firstpage

2

Abstract

Results of the ongoing medical image analysis project leading towards development of a comprehensive suite of methods and tools for quantitative pulmonary image analysis from multi-detector CT image data are presented. This includes methodology for segmentation of lung, lung lobe, bronchial tree and vascular tree, together with methods for quantitative analysis of the segmented structures. Methods for matching of structures over time that allow longitudinal comparisons of the measured quantitative indices are discussed. In addition, texture analysis and pattern recognition techniques used for tissue characterization of pulmonary parenchyma, and validation results, are reported. The future directions of quantitative and highly automated approaches to pulmonary image analysis are addressed.

Keywords

biological tissues; computerised tomography; diseases; image segmentation; image texture; lung; medical image processing; pattern recognition; bronchial tree segmentation; lung lobe segmentation; lung segmentation; multidetector CT image data; pattern recognition; pulmonary image analysis; pulmonary parenchyma; texture analysis; tissue characterization; vascular tree segmentation; Biomedical imaging; Diseases; Drugs; Educational institutions; Image analysis; Image segmentation; Image texture analysis; Life estimation; Lungs; Sensitivity and specificity;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer-Based Medical Systems, 2004. CBMS 2004. Proceedings. 17th IEEE Symposium on

ISSN

1063-7125

Print_ISBN

0-7695-2104-5

Type

conf

DOI

10.1109/CBMS.2004.1311681

Filename

1311681