Classification of gastric tumors on gastric biopsy images

Ryo Ohtsuki, Toshiyuki Tanaka

Research output: Contribution to conferencePaper

Abstract

The well-known Group Classification method for hematoxylin and eojin stained gastric tumors uses morphological features of histology patterns within a tissue slide to classify it into 5 grades from Group1 to Group5. Our approach developed an automated classification method being used for automated Group Classification of gastric tumor images. We have demonstrate the performance of the proposed method for a three class classification ± Group1 (benign), Group3 (gastric adenoma), Group5 (gastric cancer) ± on a 90 teaching dataset and 90 test dataset using Support Vector Machine and achieved accuracy of 75.6% on Group1, 64.4% on Group3, and 95.6% on Group5. Our approach combines the morphological features such as nuclear-cytoplasmic ratio, some texture features, and HLAC (higher order local autocorrelation).

Original languageEnglish
Pages2502-2506
Number of pages5
Publication statusPublished - 2013 Jan 1
Event2013 52nd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2013 - Nagoya, Japan
Duration: 2013 Sep 142013 Sep 17

Other

Other2013 52nd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2013
CountryJapan
CityNagoya
Period13/9/1413/9/17

Keywords

  • Automated classification
  • Digital pathology
  • Gastric biopsy
  • Image processing

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Classification of gastric tumors on gastric biopsy images'. Together they form a unique fingerprint.

  • Cite this

    Ohtsuki, R., & Tanaka, T. (2013). Classification of gastric tumors on gastric biopsy images. 2502-2506. Paper presented at 2013 52nd Annual Conference of the Society of Instrument and Control Engineers of Japan, SICE 2013, Nagoya, Japan.