Computational cell classification methodology for hepatocellular carcinoma

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

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)


Liver cancer is one of the frequent causes of death in the world. Hepatocellular carcinoma (HCC) is the most common histological type of primary liver cancer. HCC can be graded according to the malignancy of the tumors. Generally, a HCC grade is determined based on the characteristics of liver cell nuclei. This paper illustrates a methodology for classifying liver cell nuclei and grading HCC histological images quantitatively. The liver cell nuclei are classified in three consecutive tasks: nuclear segmentation, fibrous region detection, and nuclear classification. Each task utilizes the pixel-based textural features that are obtained through multifractal computation on digital images. First, the system segments every possible type of nuclei and excludes the nuclei within fibrous regions. Then, it classifies the rest of the nuclei to discriminate liver cell nuclei. For tumor grading, this method utilizes the following four categories of nuclear features: inner texture, geometry, spatial distribution, and surrounding texture. The proposed method was employed to classify a set of HCC histological images into five.

Original languageEnglish
Number of pages7
Publication statusPublished - 2013 Jan 1
Event2013 International Conference on Advances in ICT for Emerging Regions, ICTer 2 - Colombo, Sri Lanka
Duration: 2013 Dec 122013 Dec 13


Other2013 International Conference on Advances in ICT for Emerging Regions, ICTer 2
Country/TerritorySri Lanka


  • Cancer grading
  • Feature selection
  • HCC histological images
  • Multifractal computation
  • Multifractal measures
  • Segmentation
  • Textural feature descriptor

ASJC Scopus subject areas

  • Management of Technology and Innovation


Dive into the research topics of 'Computational cell classification methodology for hepatocellular carcinoma'. Together they form a unique fingerprint.

Cite this