Multifractal feature based cancer detection for pathological images

Chamidu Atupelage; Hiroshi Nagahashi; Michiie Sakamoto; Masahiro Yamaguchi; Akinori Hashiguchi

doi:10.1109/icbbe.2011.5780208

Multifractal feature based cancer detection for pathological images

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

Electron Microscope Laboratory

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

This paper presents a significant multifractal feature description based texture discriminating technique to examine the cancer and non-cancer regions in pathological images. We acquired the characteristics (local singularity and global regularity information) of the texture using multifractal computation and used them to discriminate the highly complex visual patterns shown in the pathological images. The proposed feature description method was applied two different samples of pathological cancer liver images in different magnifications (given by digital slider) with different patch sizes (patch is a local window that we used to capture the data for training and testing). The outcomes of the experiments indicate that the proposed multifractal feature description based texture classification method is remarkable.

Original language	English
Title of host publication	5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011
DOIs	https://doi.org/10.1109/icbbe.2011.5780208
Publication status	Published - 2011
Event	5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011 - Wuhan, China Duration: 2011 May 10 → 2011 May 12

Publication series

Name	5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011

Other

Other	5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011
Country/Territory	China
City	Wuhan
Period	11/5/10 → 11/5/12

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

Keywords

Component
Digital slider
Fractal dimension
Multifractal analysis
Pathological images
Texture classification

ASJC Scopus subject areas

Biomedical Engineering
Health Informatics
Health Information Management

Access to Document

10.1109/icbbe.2011.5780208

Cite this

Atupelage, C., Nagahashi, H., Sakamoto, M., Yamaguchi, M., & Hashiguchi, A. (2011). Multifractal feature based cancer detection for pathological images. In 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011 Article 5780208 (5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011). https://doi.org/10.1109/icbbe.2011.5780208

Multifractal feature based cancer detection for pathological images. / Atupelage, Chamidu; Nagahashi, Hiroshi; Sakamoto, Michiie et al.
5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011. 2011. 5780208 (5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Atupelage, C, Nagahashi, H, Sakamoto, M, Yamaguchi, M & Hashiguchi, A 2011, Multifractal feature based cancer detection for pathological images. in 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011., 5780208, 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011, 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011, Wuhan, China, 11/5/10. https://doi.org/10.1109/icbbe.2011.5780208

@inproceedings{1b00d5cbfced45d39a46013dfec1c780,

title = "Multifractal feature based cancer detection for pathological images",

abstract = "This paper presents a significant multifractal feature description based texture discriminating technique to examine the cancer and non-cancer regions in pathological images. We acquired the characteristics (local singularity and global regularity information) of the texture using multifractal computation and used them to discriminate the highly complex visual patterns shown in the pathological images. The proposed feature description method was applied two different samples of pathological cancer liver images in different magnifications (given by digital slider) with different patch sizes (patch is a local window that we used to capture the data for training and testing). The outcomes of the experiments indicate that the proposed multifractal feature description based texture classification method is remarkable.",

keywords = "Component, Digital slider, Fractal dimension, Multifractal analysis, Pathological images, Texture classification",

author = "Chamidu Atupelage and Hiroshi Nagahashi and Michiie Sakamoto and Masahiro Yamaguchi and Akinori Hashiguchi",

year = "2011",

doi = "10.1109/icbbe.2011.5780208",

language = "English",

isbn = "9781424450893",

series = "5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011",

booktitle = "5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011",

note = "5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011 ; Conference date: 10-05-2011 Through 12-05-2011",

}

TY - GEN

T1 - Multifractal feature based cancer detection for pathological images

AU - Atupelage, Chamidu

AU - Nagahashi, Hiroshi

AU - Sakamoto, Michiie

AU - Yamaguchi, Masahiro

AU - Hashiguchi, Akinori

PY - 2011

Y1 - 2011

N2 - This paper presents a significant multifractal feature description based texture discriminating technique to examine the cancer and non-cancer regions in pathological images. We acquired the characteristics (local singularity and global regularity information) of the texture using multifractal computation and used them to discriminate the highly complex visual patterns shown in the pathological images. The proposed feature description method was applied two different samples of pathological cancer liver images in different magnifications (given by digital slider) with different patch sizes (patch is a local window that we used to capture the data for training and testing). The outcomes of the experiments indicate that the proposed multifractal feature description based texture classification method is remarkable.

AB - This paper presents a significant multifractal feature description based texture discriminating technique to examine the cancer and non-cancer regions in pathological images. We acquired the characteristics (local singularity and global regularity information) of the texture using multifractal computation and used them to discriminate the highly complex visual patterns shown in the pathological images. The proposed feature description method was applied two different samples of pathological cancer liver images in different magnifications (given by digital slider) with different patch sizes (patch is a local window that we used to capture the data for training and testing). The outcomes of the experiments indicate that the proposed multifractal feature description based texture classification method is remarkable.

KW - Component

KW - Digital slider

KW - Fractal dimension

KW - Multifractal analysis

KW - Pathological images

KW - Texture classification

UR - https://www.scopus.com/pages/publications/79960145748

UR - https://www.scopus.com/pages/publications/79960145748#tab=citedBy

U2 - 10.1109/icbbe.2011.5780208

DO - 10.1109/icbbe.2011.5780208

M3 - Conference contribution

AN - SCOPUS:79960145748

SN - 9781424450893

T3 - 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011

BT - 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011

T2 - 5th International Conference on Bioinformatics and Biomedical Engineering, iCBBE 2011

Y2 - 10 May 2011 through 12 May 2011

ER -

Multifractal feature based cancer detection for pathological images

Abstract

Publication series

Other

UN SDGs

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this