Diagnosis of Hepatic Fibrosis by Ultrasonic Image Analysis

Author

Zhou, Guohui ; Wang, Yuanyuan ; Wang, Weiqi

Author_Institution

Dept. of Electron. Eng., Fudan Univ., Shanghai, China

fYear

2012

fDate

28-30 May 2012

Firstpage

775

Lastpage

776

Abstract

Textural features were extracted from livers´ B-mode ultrasonic images to diagnose hepatic fibrosis for chronic hepatitis-B patients (CHB). For a selected region of interest (ROI) of an ultrasonic image, 13 textural parameters based on its gray level co-occurrence matrix (GLCM) were calculated. A Fisher linear classifier was built using 83 images as its training set. The results of the test on 38 cases showed that the sensitivity of the classifier is 81.0%, and the speciality is 70.6%.

Keywords

diseases; feature extraction; image classification; image colour analysis; image texture; liver; medical image processing; ultrasonic imaging; B-mode ultrasonic images; CHB; Fisher linear classifier; GLCM; ROI; chronic hepatitis-B patients; diagnose hepatic fibrosis; gray level co-occurrence matrix; hepatic fibrosis diagnosis; livers; region of interest; textural feature extraction; ultrasonic image analysis; Acoustics; Design automation; Feature extraction; Liver; Medical services; Training; Ultrasonic imaging; Fisher linear classifier; gray level co-occurrence matrix; hepatic fibrosis; texture analysis; ultrasonic image;

fLanguage

English

Publisher

ieee

Conference_Titel

Biomedical Engineering and Biotechnology (iCBEB), 2012 International Conference on

Conference_Location

Macau, Macao

Print_ISBN

978-1-4577-1987-5

Type

conf

DOI

10.1109/iCBEB.2012.143

Filename

6245235