Three-dimensional probabilistic neural network using for MR image segmentation

Author

Yuanfeng, Lian ; Falin, Wu

Author_Institution

Coll. of Geophys. & Inf. Eng., China Univ. of Pet., Beijing, China

Volume

3

fYear

2011

Firstpage

127

Lastpage

131

Abstract

The three-dimensional probabilistic neural network (PNN) is proposed as the core classifier for segmentation of three-dimensional (3-D) magnetic resonance imaging (MRI). The proposed algorithm takes into account the spatial information between image voxels. It adopts the self-organizing map (SOM) neural network to overly segment the 3D MR image, and yield reference voxels necessary for probabilistic density function. The experimental results demonstrate the effectiveness and robustness of the proposed approach.

Keywords

biomedical MRI; image classification; image segmentation; medical image processing; probability; self-organising feature maps; 3D MR image segmentation; 3D PNN; 3D probabilistic neural network; SOM neural network; image classifier; image voxel spatial information; magnetic resonance imaging; probability density function; self organizing map; Biological neural networks; Image segmentation; Magnetic resonance imaging; Neurons; Probabilistic logic; Tensile stress; Training; MR image; segmentation; structure tensor; three-dimensional adaptive probabilistic neural network;

fLanguage

English

Publisher

ieee

Conference_Titel

Electronic Measurement & Instruments (ICEMI), 2011 10th International Conference on

Conference_Location

Chengdu

Print_ISBN

978-1-4244-8158-3

Type

conf

DOI

10.1109/ICEMI.2011.6037870

Filename

6037870