Title :
Segmentation of ultrasound images by using an incremental self-organized map
Author :
Kurnaz, M.N. ; Dokur, Z. ; Ölmez, T.
Author_Institution :
Dept. of Electron. & Commun., Istanbul Tech. Univ., Turkey
Abstract :
A new incremental self-organized map is proposed for the segmentation of the ultrasound images. Elements of the feature vectors are formed by the fast Fourier transform (FFT) of image intensities in 4×4 square blocks. In this study, two neural networks for segmentation are comparatively examined: Kohonen map, and incremental self-organized map (ISOM). It is observed that ISOM gives the best classification performance with less number of nodes after a short training time.
Keywords :
biomedical ultrasonics; fast Fourier transforms; image classification; image segmentation; medical image processing; multilayer perceptrons; self-organising feature maps; unsupervised learning; Kohonen map; classification performance; fast Fourier transform; feature vectors; image intensities; incremental self-organized map; neural networks; short training time; ultrasound image segmentation; Artificial neural networks; Biomedical imaging; Fast Fourier transforms; Feature extraction; Image segmentation; Multi-layer neural network; Neural networks; Pixel; Ultrasonic imaging; Wavelet transforms;
Conference_Titel :
Engineering in Medicine and Biology Society, 2001. Proceedings of the 23rd Annual International Conference of the IEEE
Print_ISBN :
0-7803-7211-5
DOI :
10.1109/IEMBS.2001.1017324
