New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -

Y. Tanimoto, Masanobu Takahashi, K. Oguruma, M. Nakano

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the field of histopathologic diagnosis, differential diagnosis of borderline lesions is a serious problem. Especially, differential diagnosis between early welldifferentiated hepatocellular carcinoma (ewHCC) and noncancer is difficult because the cellular atypism of ewHCC is very low. Nuclear density (number of nuclei per unit area) is one of features effective to diagnose ewHCC. In this paper, we propose new feature, degree of nuclear concentration, which represents the degree how densely nuclei are locally distributed. Two methods, counting method and density method, are proposed to quantify this feature. Counting method detects the dense regions, the regions where nuclei are densely distributed, by counting the number of nuclei in a circle. Density method converts each nuclear position to density distribution, and detects the dense regions as the regions having high density value. About 90% of correct ratio was obtained for both methods by the experiment, which shows effectiveness of this new feature. The feature was effective even if the nuclear density was normalized. Relative index, the ratio of features between ewHCC and non-cancer, was also shown to become another effective feature.

Original languageEnglish
Title of host publicationIFMBE Proceedings
Pages1083-1086
Number of pages4
Volume25
Edition4
DOIs
Publication statusPublished - 2009
EventWorld Congress on Medical Physics and Biomedical Engineering: Image Processing, Biosignal Processing, Modelling and Simulation, Biomechanics - Munich
Duration: 2009 Sep 72009 Sep 12

Other

OtherWorld Congress on Medical Physics and Biomedical Engineering: Image Processing, Biosignal Processing, Modelling and Simulation, Biomechanics
CityMunich
Period09/9/709/9/12

Fingerprint

Experiments

Keywords

  • Density
  • Hepatocellular carcinoma
  • Histopathology
  • Nuclear concentration
  • Nucleus

ASJC Scopus subject areas

  • Biomedical Engineering
  • Bioengineering

Cite this

New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -. / Tanimoto, Y.; Takahashi, Masanobu; Oguruma, K.; Nakano, M.

IFMBE Proceedings. Vol. 25 4. ed. 2009. p. 1083-1086.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tanimoto, Y, Takahashi, M, Oguruma, K & Nakano, M 2009, New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -. in IFMBE Proceedings. 4 edn, vol. 25, pp. 1083-1086, World Congress on Medical Physics and Biomedical Engineering: Image Processing, Biosignal Processing, Modelling and Simulation, Biomechanics, Munich, 09/9/7. https://doi.org/10.1007/978-3-642-03882-2-288
Tanimoto, Y. ; Takahashi, Masanobu ; Oguruma, K. ; Nakano, M. / New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -. IFMBE Proceedings. Vol. 25 4. ed. 2009. pp. 1083-1086
@inproceedings{4767dfff55324b5f9a9b880b80fd67ba,
title = "New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -",
abstract = "In the field of histopathologic diagnosis, differential diagnosis of borderline lesions is a serious problem. Especially, differential diagnosis between early welldifferentiated hepatocellular carcinoma (ewHCC) and noncancer is difficult because the cellular atypism of ewHCC is very low. Nuclear density (number of nuclei per unit area) is one of features effective to diagnose ewHCC. In this paper, we propose new feature, degree of nuclear concentration, which represents the degree how densely nuclei are locally distributed. Two methods, counting method and density method, are proposed to quantify this feature. Counting method detects the dense regions, the regions where nuclei are densely distributed, by counting the number of nuclei in a circle. Density method converts each nuclear position to density distribution, and detects the dense regions as the regions having high density value. About 90{\%} of correct ratio was obtained for both methods by the experiment, which shows effectiveness of this new feature. The feature was effective even if the nuclear density was normalized. Relative index, the ratio of features between ewHCC and non-cancer, was also shown to become another effective feature.",
keywords = "Density, Hepatocellular carcinoma, Histopathology, Nuclear concentration, Nucleus",
author = "Y. Tanimoto and Masanobu Takahashi and K. Oguruma and M. Nakano",
year = "2009",
doi = "10.1007/978-3-642-03882-2-288",
language = "English",
isbn = "9783642038815",
volume = "25",
pages = "1083--1086",
booktitle = "IFMBE Proceedings",
edition = "4",

}

TY - GEN

T1 - New feature for histopathologic diagnosis of early hepatocellular carcinoma - Degree of nuclear concentration -

AU - Tanimoto, Y.

AU - Takahashi, Masanobu

AU - Oguruma, K.

AU - Nakano, M.

PY - 2009

Y1 - 2009

N2 - In the field of histopathologic diagnosis, differential diagnosis of borderline lesions is a serious problem. Especially, differential diagnosis between early welldifferentiated hepatocellular carcinoma (ewHCC) and noncancer is difficult because the cellular atypism of ewHCC is very low. Nuclear density (number of nuclei per unit area) is one of features effective to diagnose ewHCC. In this paper, we propose new feature, degree of nuclear concentration, which represents the degree how densely nuclei are locally distributed. Two methods, counting method and density method, are proposed to quantify this feature. Counting method detects the dense regions, the regions where nuclei are densely distributed, by counting the number of nuclei in a circle. Density method converts each nuclear position to density distribution, and detects the dense regions as the regions having high density value. About 90% of correct ratio was obtained for both methods by the experiment, which shows effectiveness of this new feature. The feature was effective even if the nuclear density was normalized. Relative index, the ratio of features between ewHCC and non-cancer, was also shown to become another effective feature.

AB - In the field of histopathologic diagnosis, differential diagnosis of borderline lesions is a serious problem. Especially, differential diagnosis between early welldifferentiated hepatocellular carcinoma (ewHCC) and noncancer is difficult because the cellular atypism of ewHCC is very low. Nuclear density (number of nuclei per unit area) is one of features effective to diagnose ewHCC. In this paper, we propose new feature, degree of nuclear concentration, which represents the degree how densely nuclei are locally distributed. Two methods, counting method and density method, are proposed to quantify this feature. Counting method detects the dense regions, the regions where nuclei are densely distributed, by counting the number of nuclei in a circle. Density method converts each nuclear position to density distribution, and detects the dense regions as the regions having high density value. About 90% of correct ratio was obtained for both methods by the experiment, which shows effectiveness of this new feature. The feature was effective even if the nuclear density was normalized. Relative index, the ratio of features between ewHCC and non-cancer, was also shown to become another effective feature.

KW - Density

KW - Hepatocellular carcinoma

KW - Histopathology

KW - Nuclear concentration

KW - Nucleus

UR - http://www.scopus.com/inward/record.url?scp=77950170498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77950170498&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-03882-2-288

DO - 10.1007/978-3-642-03882-2-288

M3 - Conference contribution

SN - 9783642038815

VL - 25

SP - 1083

EP - 1086

BT - IFMBE Proceedings

ER -