Self-organizing map for segmenting 3D biological images

An image processing method for features extraction and segmentation from three-dimensional (3D) image datasets is presented. Kohonen's self-organizing map (SOM) is used to perform segmentation. Previously, the segmentation method worked on a 2D dataset based on a projection of the three-dimensional dataset (Nguyen et al., 1998). Our 3D approach to segment biological images preserves the 3D object orientations with respect to the surrounding cell volume. A few examples from genetics and brain analysis are provided in order to demonstrate the performance of the proposed method.

[1]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[2]  Partha Pratim Das,et al.  Segmentation of three-dimensional surfaces , 1990, Pattern Recognit. Lett..

[3]  Juha Ylä-Jääski,et al.  Segmentation and analysis of 3D volume images , 1988, [1988 Proceedings] 9th International Conference on Pattern Recognition.

[4]  George Papadourakis,et al.  Object recognition using invariant object boundary representations and neural network models , 1992, Pattern Recognit..

[5]  R. Mehrotra,et al.  A two-stage neural net for segmentation of range images , 1993, IEEE International Conference on Neural Networks.

[6]  Andreas Pommert,et al.  3D-segmentation and display of tomographic imagery , 1988, [1988 Proceedings] 9th International Conference on Pattern Recognition.