Classification of cirrhotic liver on MR images using texture analysis

G. N. Lee; X. Zhang; M. Kaneniatsu; X. Zhou; T. Hara; H. Kato; H. Kondo; H. Fujita; H. Hoshi

Classification of cirrhotic liver on MR images using texture analysis

G. N. Lee
, X. Zhang
, M. Kaneniatsu
, X. Zhou
, T. Hara
, H. Kato
, H. Kondo
, H. Fujita
, H. Hoshi

Gifu University

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

5 Scopus citations

Abstract

A total of 128 features, including 126 co-occurrence matrix based texture features and two statistical measures, are employed for the classification of cirrhotic and non-cirrhotic liver A genetic algorithm is used for selecting the feature subsets that have the most discriminative power. A three-layer back-propagation neural network is used for the classification. Based on a training set of 15 cirrhotic cases and 30 non-cirrhotic cases, the best testing Az is 0.73 obtained by using six features.

Original language	English
Pages (from-to)	379-381
Number of pages	3
Journal	International journal of computer assisted radiology and surgery
Volume	1
Issue number	SUPPL. 7
State	Published - Jun 2006
Externally published	Yes

Keywords

Artificial neural network
Cirrhosis of the liver (abdominal)
Computer-aided diagnosis
MR imaging
Texture analysis

Cite this

@article{046d41569e264dadbbe3b2a6984c88f5,

title = "Classification of cirrhotic liver on MR images using texture analysis",

abstract = "A total of 128 features, including 126 co-occurrence matrix based texture features and two statistical measures, are employed for the classification of cirrhotic and non-cirrhotic liver A genetic algorithm is used for selecting the feature subsets that have the most discriminative power. A three-layer back-propagation neural network is used for the classification. Based on a training set of 15 cirrhotic cases and 30 non-cirrhotic cases, the best testing Az is 0.73 obtained by using six features.",

keywords = "Artificial neural network, Cirrhosis of the liver (abdominal), Computer-aided diagnosis, MR imaging, Texture analysis",

author = "Lee, \{G. N.\} and X. Zhang and M. Kaneniatsu and X. Zhou and T. Hara and H. Kato and H. Kondo and H. Fujita and H. Hoshi",

year = "2006",

month = jun,

language = "英語",

volume = "1",

pages = "379--381",

journal = "International journal of computer assisted radiology and surgery",

issn = "1861-6410",

publisher = "Springer Science and Business Media Deutschland GmbH",

number = "SUPPL. 7",

}

TY - JOUR

T1 - Classification of cirrhotic liver on MR images using texture analysis

AU - Lee, G. N.

AU - Zhang, X.

AU - Kaneniatsu, M.

AU - Zhou, X.

AU - Hara, T.

AU - Kato, H.

AU - Kondo, H.

AU - Fujita, H.

AU - Hoshi, H.

PY - 2006/6

Y1 - 2006/6

N2 - A total of 128 features, including 126 co-occurrence matrix based texture features and two statistical measures, are employed for the classification of cirrhotic and non-cirrhotic liver A genetic algorithm is used for selecting the feature subsets that have the most discriminative power. A three-layer back-propagation neural network is used for the classification. Based on a training set of 15 cirrhotic cases and 30 non-cirrhotic cases, the best testing Az is 0.73 obtained by using six features.

AB - A total of 128 features, including 126 co-occurrence matrix based texture features and two statistical measures, are employed for the classification of cirrhotic and non-cirrhotic liver A genetic algorithm is used for selecting the feature subsets that have the most discriminative power. A three-layer back-propagation neural network is used for the classification. Based on a training set of 15 cirrhotic cases and 30 non-cirrhotic cases, the best testing Az is 0.73 obtained by using six features.

KW - Artificial neural network

KW - Cirrhosis of the liver (abdominal)

KW - Computer-aided diagnosis

KW - MR imaging

KW - Texture analysis

UR - https://www.scopus.com/pages/publications/33748954792

M3 - 記事

AN - SCOPUS:33748954792

SN - 1861-6410

VL - 1

SP - 379

EP - 381

JO - International journal of computer assisted radiology and surgery

JF - International journal of computer assisted radiology and surgery

IS - SUPPL. 7

ER -

Classification of cirrhotic liver on MR images using texture analysis

Abstract

Keywords

Other files and links

Fingerprint

Cite this