Neural networks for model-based segmentation of MR brain images

Author

Gindi, Gene ; Rangarajan, Anand ; Zubal, I. George

Author_Institution

Dept. of Radiol., State Univ. of New York, Stony Brook, NY, USA

fYear

1993

fDate

1993

Firstpage

90

Lastpage

92

Abstract

Automated segmentation of magnetic resonance (MR) brain imagery into anatomical regions is a complex task that needs contextual guidance to overcome problems associated with noise, missing data, and the overlap of features associated with different anatomical regions. In this work, the contextual information is provided as an anatomical brain atlas. The matching of atlas to image data is represented by a set of deformable contours that seek compromise fits between expected model information and image data.

Keywords

biomedical NMR; brain; image segmentation; medical image processing; neural nets; MR brain images; anatomical brain atlas; anatomical regions; automated segmentation; contextual guidance; deformable contours; features overlap; magnetic resonance imaging; medical diagnostic imaging; missing data; model-based segmentation; noise; Biological neural networks; Biomedical imaging; Brain modeling; Computed tomography; Computer science; Image segmentation; Magnetic resonance; Medical diagnostic imaging; Radiology; Surgery;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Engineering Conference, 1993., Proceedings of the Twelfth Southern

Conference_Location

New Orleans, LA, USA

Print_ISBN

0-7803-0976-6

Type

conf

DOI

10.1109/SBEC.1993.247341

Filename

247341