Segmentation of ultrasound images by using an incremental self-organized map

M. N. Kurnaz; Z. Dokur; T. Ölmez

Segmentation of ultrasound images by using an incremental self-organized map

M. N. Kurnaz^*, Z. Dokur, T. Ölmez

^*Corresponding author for this work

Department of Electronics and Communication Engineering

Istanbul Technical University

Research output: Contribution to journal › Conference article › peer-review

8 Citations (Scopus)

Abstract

This paper presents a new segmentation method for ultrasound images. 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.

Original language	English
Pages (from-to)	2638-2640
Number of pages	3
Journal	Annual Reports of the Research Reactor Institute, Kyoto University
Volume	3
Publication status	Published - 2001
Event	23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society - Istanbul, Turkey Duration: 25 Oct 2001 → 28 Oct 2001

Keywords

Classification
Multilayer perceptron
Neural networks
Segmentation of biomedical images
Self organized map

Cite this

@article{0aed86c261794be084fc5e0018e04f66,

title = "Segmentation of ultrasound images by using an incremental self-organized map",

abstract = "This paper presents a new segmentation method for ultrasound images. 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 = "Classification, Multilayer perceptron, Neural networks, Segmentation of biomedical images, Self organized map",

author = "Kurnaz, \{M. N.\} and Z. Dokur and T. {\"O}lmez",

year = "2001",

language = "English",

volume = "3",

pages = "2638--2640",

journal = "Annual Reports of the Research Reactor Institute, Kyoto University",

issn = "0454-9244",

note = "23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society ; Conference date: 25-10-2001 Through 28-10-2001",

}

TY - JOUR

T1 - Segmentation of ultrasound images by using an incremental self-organized map

AU - Kurnaz, M. N.

AU - Dokur, Z.

AU - Ölmez, T.

PY - 2001

Y1 - 2001

N2 - This paper presents a new segmentation method for ultrasound images. 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.

AB - This paper presents a new segmentation method for ultrasound images. 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.

KW - Classification

KW - Multilayer perceptron

KW - Neural networks

KW - Segmentation of biomedical images

KW - Self organized map

UR - https://www.scopus.com/pages/publications/0035782015

M3 - Conference article

AN - SCOPUS:0035782015

SN - 0454-9244

VL - 3

SP - 2638

EP - 2640

JO - Annual Reports of the Research Reactor Institute, Kyoto University

JF - Annual Reports of the Research Reactor Institute, Kyoto University

T2 - 23rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society

Y2 - 25 October 2001 through 28 October 2001

ER -

Segmentation of ultrasound images by using an incremental self-organized map

Abstract

Keywords

Other files and links

Fingerprint

Cite this