3D segmentation and labeling using self-organizing Kohonen network for volumetric measurements on brain CT imaging to quantify TBI recovery

Author

Ahmed, Mohamed N. ; Farag, Aly A.

Author_Institution

Dept. of Electr. Eng., Louisville Univ., KY, USA

Volume

2

fYear

1996

fDate

31 Oct-3 Nov 1996

Firstpage

738

Abstract

In this paper, we present a new system to segment and label CT brain slices using a self-organizing Kohonen network. Our aim is to extract reliable and robust measures from CT images of Traumatic Brain Injury (TBI) patients that can accurately describe the morphological changes in the brain as recovery progresses. Segmentation is performed by assigning a feature pattern to each voxel, consisting of a scaled family of differential geometrical invariant features. The invariant feature pattern is input to Kohonen network for an unsupervised classification of the voxels into regions

Keywords

biomedical NMR; brain; computerised tomography; image classification; image segmentation; medical expert systems; medical image processing; self-organising feature maps; unsupervised learning; 3D segmentation; Laplacian of Gaussian operator; brain computed tomography imaging; brain slice labeling; differential geometrical invariant features; feature pattern; morphological changes; neural nets; scaled family; self-organizing Kohonen network; traumatic brain injury recovery; unsupervised classification; volumetric measurements; voxels into regions; Brain injuries; Computed tomography; Data mining; Electric variables measurement; Engineering in Medicine and Biology Society; Filters; Image segmentation; Labeling; Layout; Volume measurement;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE

Conference_Location

Amsterdam

Print_ISBN

0-7803-3811-1

Type

conf

DOI

10.1109/IEMBS.1996.651953

Filename

651953