Texture analysis and classification of ultrasound liver images

Shuang Gao; Yuhua Peng; Huizhi Guo; Weifeng Liu; Tianxin Gao; Yuanqing Xu; Xiaoying Tang

doi:10.3233/BME-130922

Texture analysis and classification of ultrasound liver images

Shuang Gao, Yuhua Peng, Huizhi Guo, Weifeng Liu, Tianxin Gao, Yuanqing Xu, Xiaoying Tang^*

^*Corresponding author for this work

School of Medical and Technology

Beijing Institute of Technology

Research output: Contribution to journal › Article › peer-review

52 Citations (Scopus)

Abstract

Ultrasound as a noninvasive imaging technique is widely used to diagnose liver diseases. Texture analysis and classification of ultrasound liver images have become an important research topic across the world. In this study, GLGCM (Gray Level Gradient Co-Occurrence Matrix) was implemented for texture analysis of ultrasound liver images first, followed by the use of GLCM (Gray Level Co-occurrence Matrix) at the second stage. Twenty two features were obtained using the two methods, andseven most powerful features were selected for classification using BP (Back Propagation) neural network. Fibrosis was divided into five stages (S0-S4) in this study. The classification accuracies of S0-S4 were 100%, 90%, 70%, 90% and 100%, respectively.

Original language	English
Pages (from-to)	1209-1216
Number of pages	8
Journal	Bio-Medical Materials and Engineering
Volume	24
Issue number	1
DOIs	https://doi.org/10.3233/BME-130922
Publication status	Published - 2014

Keywords

artificial neural network
liver fibrosis
texture features analysis
texture features extraction
ultrasonic image

Access to Document

10.3233/BME-130922

Cite this

@article{156212cebe2f4a1f80bdd0378fe4fbc6,

title = "Texture analysis and classification of ultrasound liver images",

abstract = "Ultrasound as a noninvasive imaging technique is widely used to diagnose liver diseases. Texture analysis and classification of ultrasound liver images have become an important research topic across the world. In this study, GLGCM (Gray Level Gradient Co-Occurrence Matrix) was implemented for texture analysis of ultrasound liver images first, followed by the use of GLCM (Gray Level Co-occurrence Matrix) at the second stage. Twenty two features were obtained using the two methods, andseven most powerful features were selected for classification using BP (Back Propagation) neural network. Fibrosis was divided into five stages (S0-S4) in this study. The classification accuracies of S0-S4 were 100%, 90%, 70%, 90% and 100%, respectively.",

keywords = "artificial neural network, liver fibrosis, texture features analysis, texture features extraction, ultrasonic image",

author = "Shuang Gao and Yuhua Peng and Huizhi Guo and Weifeng Liu and Tianxin Gao and Yuanqing Xu and Xiaoying Tang",

year = "2014",

doi = "10.3233/BME-130922",

language = "English",

volume = "24",

pages = "1209--1216",

journal = "Bio-Medical Materials and Engineering",

issn = "0959-2989",

publisher = "IOS Press BV",

number = "1",

}

TY - JOUR

T1 - Texture analysis and classification of ultrasound liver images

AU - Gao, Shuang

AU - Peng, Yuhua

AU - Guo, Huizhi

AU - Liu, Weifeng

AU - Gao, Tianxin

AU - Xu, Yuanqing

AU - Tang, Xiaoying

PY - 2014

Y1 - 2014

N2 - Ultrasound as a noninvasive imaging technique is widely used to diagnose liver diseases. Texture analysis and classification of ultrasound liver images have become an important research topic across the world. In this study, GLGCM (Gray Level Gradient Co-Occurrence Matrix) was implemented for texture analysis of ultrasound liver images first, followed by the use of GLCM (Gray Level Co-occurrence Matrix) at the second stage. Twenty two features were obtained using the two methods, andseven most powerful features were selected for classification using BP (Back Propagation) neural network. Fibrosis was divided into five stages (S0-S4) in this study. The classification accuracies of S0-S4 were 100%, 90%, 70%, 90% and 100%, respectively.

AB - Ultrasound as a noninvasive imaging technique is widely used to diagnose liver diseases. Texture analysis and classification of ultrasound liver images have become an important research topic across the world. In this study, GLGCM (Gray Level Gradient Co-Occurrence Matrix) was implemented for texture analysis of ultrasound liver images first, followed by the use of GLCM (Gray Level Co-occurrence Matrix) at the second stage. Twenty two features were obtained using the two methods, andseven most powerful features were selected for classification using BP (Back Propagation) neural network. Fibrosis was divided into five stages (S0-S4) in this study. The classification accuracies of S0-S4 were 100%, 90%, 70%, 90% and 100%, respectively.

KW - artificial neural network

KW - liver fibrosis

KW - texture features analysis

KW - texture features extraction

KW - ultrasonic image

UR - http://www.scopus.com/inward/record.url?scp=84891105510&partnerID=8YFLogxK

U2 - 10.3233/BME-130922

DO - 10.3233/BME-130922

M3 - Article

C2 - 24212015

AN - SCOPUS:84891105510

SN - 0959-2989

VL - 24

SP - 1209

EP - 1216

JO - Bio-Medical Materials and Engineering

JF - Bio-Medical Materials and Engineering

IS - 1

ER -

Texture analysis and classification of ultrasound liver images

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this