Random patch probabilistic density algorithm for tissue localization from the whole slide images

Research output: Contribution to journalArticle

Abstract

Objects localization from the whole slide images is one main issues that can be handled by image processing techniques which is important for both, the medical and computer science fields. In this study, random patch probabilistic density method is proposed for localization the tissue from the whole slide histology images. The proposed method is a simple localization method, it based on foreground density feature inside a virtual box and the box represents a randomly selected region from the foreground of the image. The proposedmethod then used to localize the tissue from the whole slide image which represents the Region of Interest (ROI) in this case. In medical imaging, in the process of the analyzing the whole slide tissue microscopic images which is considered an extremely important step in histopathological image analysis and diagnosing, this analysis includes localizing the tissue region from the whole slide in a bounding box, before the scanning process starting. The proposed method able to localize the objects adaptively without predetermining the number of objects or clusters to be found as in K-means and Fuzzy C-means. After that, the proposed method, results, evaluation and comparison are explained.

Original languageEnglish
Pages (from-to)270-282
Number of pages13
JournalJournal of Medical Sciences (Faisalabad)
Volume14
Issue number6-8
DOIs
Publication statusPublished - 2014

Fingerprint

Medical Informatics
Diagnostic Imaging
Histology

Keywords

  • Density feature
  • Localization
  • Possible objects
  • ROI
  • Tissue
  • Whole slide tissue

ASJC Scopus subject areas

  • Medicine(all)

Cite this

@article{2480ccba3b3740b3bf2c7c09907fc572,
title = "Random patch probabilistic density algorithm for tissue localization from the whole slide images",
abstract = "Objects localization from the whole slide images is one main issues that can be handled by image processing techniques which is important for both, the medical and computer science fields. In this study, random patch probabilistic density method is proposed for localization the tissue from the whole slide histology images. The proposed method is a simple localization method, it based on foreground density feature inside a virtual box and the box represents a randomly selected region from the foreground of the image. The proposedmethod then used to localize the tissue from the whole slide image which represents the Region of Interest (ROI) in this case. In medical imaging, in the process of the analyzing the whole slide tissue microscopic images which is considered an extremely important step in histopathological image analysis and diagnosing, this analysis includes localizing the tissue region from the whole slide in a bounding box, before the scanning process starting. The proposed method able to localize the objects adaptively without predetermining the number of objects or clusters to be found as in K-means and Fuzzy C-means. After that, the proposed method, results, evaluation and comparison are explained.",
keywords = "Density feature, Localization, Possible objects, ROI, Tissue, Whole slide tissue",
author = "Aloman, {Yazan M.} and {Sheikh Abdullah}, {Siti Norul Huda} and {Md Zain}, {Reena Rahayu} and Khairuddin Omar",
year = "2014",
doi = "10.3923/jms.2014.270.282",
language = "English",
volume = "14",
pages = "270--282",
journal = "Journal of Medical Sciences (Faisalabad)",
issn = "1682-4474",
publisher = "Asian Network for Scientific Information",
number = "6-8",

}

TY - JOUR

T1 - Random patch probabilistic density algorithm for tissue localization from the whole slide images

AU - Aloman, Yazan M.

AU - Sheikh Abdullah, Siti Norul Huda

AU - Md Zain, Reena Rahayu

AU - Omar, Khairuddin

PY - 2014

Y1 - 2014

N2 - Objects localization from the whole slide images is one main issues that can be handled by image processing techniques which is important for both, the medical and computer science fields. In this study, random patch probabilistic density method is proposed for localization the tissue from the whole slide histology images. The proposed method is a simple localization method, it based on foreground density feature inside a virtual box and the box represents a randomly selected region from the foreground of the image. The proposedmethod then used to localize the tissue from the whole slide image which represents the Region of Interest (ROI) in this case. In medical imaging, in the process of the analyzing the whole slide tissue microscopic images which is considered an extremely important step in histopathological image analysis and diagnosing, this analysis includes localizing the tissue region from the whole slide in a bounding box, before the scanning process starting. The proposed method able to localize the objects adaptively without predetermining the number of objects or clusters to be found as in K-means and Fuzzy C-means. After that, the proposed method, results, evaluation and comparison are explained.

AB - Objects localization from the whole slide images is one main issues that can be handled by image processing techniques which is important for both, the medical and computer science fields. In this study, random patch probabilistic density method is proposed for localization the tissue from the whole slide histology images. The proposed method is a simple localization method, it based on foreground density feature inside a virtual box and the box represents a randomly selected region from the foreground of the image. The proposedmethod then used to localize the tissue from the whole slide image which represents the Region of Interest (ROI) in this case. In medical imaging, in the process of the analyzing the whole slide tissue microscopic images which is considered an extremely important step in histopathological image analysis and diagnosing, this analysis includes localizing the tissue region from the whole slide in a bounding box, before the scanning process starting. The proposed method able to localize the objects adaptively without predetermining the number of objects or clusters to be found as in K-means and Fuzzy C-means. After that, the proposed method, results, evaluation and comparison are explained.

KW - Density feature

KW - Localization

KW - Possible objects

KW - ROI

KW - Tissue

KW - Whole slide tissue

UR - http://www.scopus.com/inward/record.url?scp=84934279304&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84934279304&partnerID=8YFLogxK

U2 - 10.3923/jms.2014.270.282

DO - 10.3923/jms.2014.270.282

M3 - Article

VL - 14

SP - 270

EP - 282

JO - Journal of Medical Sciences (Faisalabad)

JF - Journal of Medical Sciences (Faisalabad)

SN - 1682-4474

IS - 6-8

ER -