Computational grading of hepatocellular carcinoma using multifractal feature description

Chamidu Atupelage, Hiroshi Nagahashi, Masahiro Yamaguchi, Tokiya Abe, Akinori Hashiguchi, Michiie Sakamoto

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Cancer grading has become an important topic in the field of image interpretation-based computer aided diagnosis systems. This paper proposes a novel feature descriptor to observe the characteristics of histopathological textures in a discriminative manner. The proposed feature descriptor utilizes fractal geometric analysis with four multifractal measures to construct an eight dimensional feature space. The proposed method employed a bag-of-feature-based classification model to discriminate a set of hepatocellular carcinoma images into five categories according to Edmondson and Steiner's grading system. Three feature selection methods were utilized to obtain the most discriminative features of codeword dictionary (codebook). Furthermore, we incorporated four other textural feature descriptors: Gabor-filters, LM-filters, local binary patterns, and Haralick, to obtain a benchmark of the accuracy of the classification. Two experiments were performed: (i) classifying non-neoplastic tissues and tumors and (ii) grading the hepatocellular carcinoma images into five classes. Experimental results indicated the significance of the multifractal features for describing the histopathological image texture because it outperformed other four feature descriptors. We graded a given ROI image by defining a threshold-based majority-voting rule and obtained an average correct classification rate around 95% for five classes classification.

Original languageEnglish
Pages (from-to)61-71
Number of pages11
JournalComputerized Medical Imaging and Graphics
Volume37
Issue number1
DOIs
Publication statusPublished - 2013 Jan

Fingerprint

Hepatocellular Carcinoma
Image texture
Computer aided diagnosis
Benchmarking
Gabor filters
Fractals
Neoplasm Grading
Politics
Glossaries
Feature extraction
Tumors
Textures
Tissue
Neoplasms
Experiments

Keywords

  • Cancer grading
  • Feature selection
  • Fractal geometry
  • HCC histopathological images
  • Multifractal computation
  • Multifractal measures
  • Textural feature descriptor

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Health Informatics
  • Radiological and Ultrasound Technology
  • Computer Graphics and Computer-Aided Design
  • Computer Vision and Pattern Recognition

Cite this

Computational grading of hepatocellular carcinoma using multifractal feature description. / Atupelage, Chamidu; Nagahashi, Hiroshi; Yamaguchi, Masahiro; Abe, Tokiya; Hashiguchi, Akinori; Sakamoto, Michiie.

In: Computerized Medical Imaging and Graphics, Vol. 37, No. 1, 01.2013, p. 61-71.

Research output: Contribution to journalArticle

Atupelage, Chamidu ; Nagahashi, Hiroshi ; Yamaguchi, Masahiro ; Abe, Tokiya ; Hashiguchi, Akinori ; Sakamoto, Michiie. / Computational grading of hepatocellular carcinoma using multifractal feature description. In: Computerized Medical Imaging and Graphics. 2013 ; Vol. 37, No. 1. pp. 61-71.
@article{10b5df4b5f3242a5b7a38c36ab0d9bbf,
title = "Computational grading of hepatocellular carcinoma using multifractal feature description",
abstract = "Cancer grading has become an important topic in the field of image interpretation-based computer aided diagnosis systems. This paper proposes a novel feature descriptor to observe the characteristics of histopathological textures in a discriminative manner. The proposed feature descriptor utilizes fractal geometric analysis with four multifractal measures to construct an eight dimensional feature space. The proposed method employed a bag-of-feature-based classification model to discriminate a set of hepatocellular carcinoma images into five categories according to Edmondson and Steiner's grading system. Three feature selection methods were utilized to obtain the most discriminative features of codeword dictionary (codebook). Furthermore, we incorporated four other textural feature descriptors: Gabor-filters, LM-filters, local binary patterns, and Haralick, to obtain a benchmark of the accuracy of the classification. Two experiments were performed: (i) classifying non-neoplastic tissues and tumors and (ii) grading the hepatocellular carcinoma images into five classes. Experimental results indicated the significance of the multifractal features for describing the histopathological image texture because it outperformed other four feature descriptors. We graded a given ROI image by defining a threshold-based majority-voting rule and obtained an average correct classification rate around 95{\%} for five classes classification.",
keywords = "Cancer grading, Feature selection, Fractal geometry, HCC histopathological images, Multifractal computation, Multifractal measures, Textural feature descriptor",
author = "Chamidu Atupelage and Hiroshi Nagahashi and Masahiro Yamaguchi and Tokiya Abe and Akinori Hashiguchi and Michiie Sakamoto",
year = "2013",
month = "1",
doi = "10.1016/j.compmedimag.2012.10.001",
language = "English",
volume = "37",
pages = "61--71",
journal = "Computerized Medical Imaging and Graphics",
issn = "0895-6111",
publisher = "Elsevier Limited",
number = "1",

}

TY - JOUR

T1 - Computational grading of hepatocellular carcinoma using multifractal feature description

AU - Atupelage, Chamidu

AU - Nagahashi, Hiroshi

AU - Yamaguchi, Masahiro

AU - Abe, Tokiya

AU - Hashiguchi, Akinori

AU - Sakamoto, Michiie

PY - 2013/1

Y1 - 2013/1

N2 - Cancer grading has become an important topic in the field of image interpretation-based computer aided diagnosis systems. This paper proposes a novel feature descriptor to observe the characteristics of histopathological textures in a discriminative manner. The proposed feature descriptor utilizes fractal geometric analysis with four multifractal measures to construct an eight dimensional feature space. The proposed method employed a bag-of-feature-based classification model to discriminate a set of hepatocellular carcinoma images into five categories according to Edmondson and Steiner's grading system. Three feature selection methods were utilized to obtain the most discriminative features of codeword dictionary (codebook). Furthermore, we incorporated four other textural feature descriptors: Gabor-filters, LM-filters, local binary patterns, and Haralick, to obtain a benchmark of the accuracy of the classification. Two experiments were performed: (i) classifying non-neoplastic tissues and tumors and (ii) grading the hepatocellular carcinoma images into five classes. Experimental results indicated the significance of the multifractal features for describing the histopathological image texture because it outperformed other four feature descriptors. We graded a given ROI image by defining a threshold-based majority-voting rule and obtained an average correct classification rate around 95% for five classes classification.

AB - Cancer grading has become an important topic in the field of image interpretation-based computer aided diagnosis systems. This paper proposes a novel feature descriptor to observe the characteristics of histopathological textures in a discriminative manner. The proposed feature descriptor utilizes fractal geometric analysis with four multifractal measures to construct an eight dimensional feature space. The proposed method employed a bag-of-feature-based classification model to discriminate a set of hepatocellular carcinoma images into five categories according to Edmondson and Steiner's grading system. Three feature selection methods were utilized to obtain the most discriminative features of codeword dictionary (codebook). Furthermore, we incorporated four other textural feature descriptors: Gabor-filters, LM-filters, local binary patterns, and Haralick, to obtain a benchmark of the accuracy of the classification. Two experiments were performed: (i) classifying non-neoplastic tissues and tumors and (ii) grading the hepatocellular carcinoma images into five classes. Experimental results indicated the significance of the multifractal features for describing the histopathological image texture because it outperformed other four feature descriptors. We graded a given ROI image by defining a threshold-based majority-voting rule and obtained an average correct classification rate around 95% for five classes classification.

KW - Cancer grading

KW - Feature selection

KW - Fractal geometry

KW - HCC histopathological images

KW - Multifractal computation

KW - Multifractal measures

KW - Textural feature descriptor

UR - http://www.scopus.com/inward/record.url?scp=84875372492&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875372492&partnerID=8YFLogxK

U2 - 10.1016/j.compmedimag.2012.10.001

DO - 10.1016/j.compmedimag.2012.10.001

M3 - Article

C2 - 23141965

AN - SCOPUS:84875372492

VL - 37

SP - 61

EP - 71

JO - Computerized Medical Imaging and Graphics

JF - Computerized Medical Imaging and Graphics

SN - 0895-6111

IS - 1

ER -