An application of image matching/fusion in medicine

Author

Vajdic, Steven M. ; Graham, Jeremy ; Voigt, Shaun

Author_Institution

Integra Australia Pty Ltd., The Levels, SA, Australia

fYear

1996

fDate

21-22 Nov 1996

Firstpage

42

Lastpage

47

Abstract

A 3D multimodal image matching/fusion algorithm is implemented as an automatic test bed approach for getting better image clarity and higher application speed. It is implemented onto 3D medical image data sets and based on segmentation/classification of anatomical regions of human organs. The reference and target 3D sets are segmented simultaneously into clusters with similar voxel intensities. The transformation for image matching is found using rule based analysis and similarities of segmented classes, then the sets are fused using interpolation of weighted voxel attributes

Keywords

image classification; image matching; image segmentation; interpolation; medical image processing; sensor fusion; 3D medical image data sets; 3D multimodal image fusion algorithm; anatomical regions; automatic test bed approach; classification; clusters; human organs; image clarity; image matching; interpolation; segmentation; voxel intensity; weighted voxel attributes; Biomedical engineering; Biomedical imaging; Clustering algorithms; Computed tomography; Humans; Image matching; Image segmentation; Magnetic resonance imaging; Positron emission tomography; Technological innovation;

fLanguage

English

Publisher

ieee

Conference_Titel

Data Fusion Symposium, 1996. ADFS '96., First Australian

Conference_Location

Adelaide, SA

Print_ISBN

0-7803-3601-1

Type

conf

DOI

10.1109/ADFS.1996.581079

Filename

581079