Batch-invariant nuclear segmentation in whole mount histology sections

Hang Chang, Leandro A. Loss, Paul T. Spellman, Alexander D Borowsky, Bahram Parvin

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

8 Citations (Scopus)

Abstract

The Cancer Genome Atlas (TCGA) provides a rich repository of whole mount tumor sections that are collected from different laboratories. However, there are a significant amount of technical and biological variations that impede analysis. We have developed a novel approach for nuclear segmentation in histology sections, which addresses the problem of technical and biological variations by incorporating information from manually annotated reference patches with the local color space of the original image. Subsequently, the problem is formulated within a multi-reference graph cut with geodesic constraints. This approach has been validated on manually curated samples and then applied to a dataset of 440 whole mount tissue sections, originating from different laboratories, which are typically 40k-by-40k pixels or larger. Segmentation results, through a zoomable interface, and extracted morphometric data are available at: http://tcga.lbl.gov.

Original languageEnglish (US)
Title of host publicationProceedings - International Symposium on Biomedical Imaging
Pages856-859
Number of pages4
DOIs
StatePublished - 2012
Event2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012 - Barcelona, Spain
Duration: May 2 2012May 5 2012

Other

Other2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012
CountrySpain
CityBarcelona
Period5/2/125/5/12

Fingerprint

Histology
Atlases
Tumors
Neoplasms
Color
Genes
Pixels
Genome
Tissue
Datasets

Keywords

  • H&E tissue section
  • Nuclear segmentation
  • Nuclear/Background classification

ASJC Scopus subject areas

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Cite this

Chang, H., Loss, L. A., Spellman, P. T., Borowsky, A. D., & Parvin, B. (2012). Batch-invariant nuclear segmentation in whole mount histology sections. In Proceedings - International Symposium on Biomedical Imaging (pp. 856-859). [6235683] https://doi.org/10.1109/ISBI.2012.6235683

Batch-invariant nuclear segmentation in whole mount histology sections. / Chang, Hang; Loss, Leandro A.; Spellman, Paul T.; Borowsky, Alexander D; Parvin, Bahram.

Proceedings - International Symposium on Biomedical Imaging. 2012. p. 856-859 6235683.

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

Chang, H, Loss, LA, Spellman, PT, Borowsky, AD & Parvin, B 2012, Batch-invariant nuclear segmentation in whole mount histology sections. in Proceedings - International Symposium on Biomedical Imaging., 6235683, pp. 856-859, 2012 9th IEEE International Symposium on Biomedical Imaging: From Nano to Macro, ISBI 2012, Barcelona, Spain, 5/2/12. https://doi.org/10.1109/ISBI.2012.6235683
Chang H, Loss LA, Spellman PT, Borowsky AD, Parvin B. Batch-invariant nuclear segmentation in whole mount histology sections. In Proceedings - International Symposium on Biomedical Imaging. 2012. p. 856-859. 6235683 https://doi.org/10.1109/ISBI.2012.6235683
Chang, Hang ; Loss, Leandro A. ; Spellman, Paul T. ; Borowsky, Alexander D ; Parvin, Bahram. / Batch-invariant nuclear segmentation in whole mount histology sections. Proceedings - International Symposium on Biomedical Imaging. 2012. pp. 856-859
@inproceedings{ab7da271d4e04a4eb26a6158b7ecb7dc,
title = "Batch-invariant nuclear segmentation in whole mount histology sections",
abstract = "The Cancer Genome Atlas (TCGA) provides a rich repository of whole mount tumor sections that are collected from different laboratories. However, there are a significant amount of technical and biological variations that impede analysis. We have developed a novel approach for nuclear segmentation in histology sections, which addresses the problem of technical and biological variations by incorporating information from manually annotated reference patches with the local color space of the original image. Subsequently, the problem is formulated within a multi-reference graph cut with geodesic constraints. This approach has been validated on manually curated samples and then applied to a dataset of 440 whole mount tissue sections, originating from different laboratories, which are typically 40k-by-40k pixels or larger. Segmentation results, through a zoomable interface, and extracted morphometric data are available at: http://tcga.lbl.gov.",
keywords = "H&E tissue section, Nuclear segmentation, Nuclear/Background classification",
author = "Hang Chang and Loss, {Leandro A.} and Spellman, {Paul T.} and Borowsky, {Alexander D} and Bahram Parvin",
year = "2012",
doi = "10.1109/ISBI.2012.6235683",
language = "English (US)",
isbn = "9781457718588",
pages = "856--859",
booktitle = "Proceedings - International Symposium on Biomedical Imaging",

}

TY - GEN

T1 - Batch-invariant nuclear segmentation in whole mount histology sections

AU - Chang, Hang

AU - Loss, Leandro A.

AU - Spellman, Paul T.

AU - Borowsky, Alexander D

AU - Parvin, Bahram

PY - 2012

Y1 - 2012

N2 - The Cancer Genome Atlas (TCGA) provides a rich repository of whole mount tumor sections that are collected from different laboratories. However, there are a significant amount of technical and biological variations that impede analysis. We have developed a novel approach for nuclear segmentation in histology sections, which addresses the problem of technical and biological variations by incorporating information from manually annotated reference patches with the local color space of the original image. Subsequently, the problem is formulated within a multi-reference graph cut with geodesic constraints. This approach has been validated on manually curated samples and then applied to a dataset of 440 whole mount tissue sections, originating from different laboratories, which are typically 40k-by-40k pixels or larger. Segmentation results, through a zoomable interface, and extracted morphometric data are available at: http://tcga.lbl.gov.

AB - The Cancer Genome Atlas (TCGA) provides a rich repository of whole mount tumor sections that are collected from different laboratories. However, there are a significant amount of technical and biological variations that impede analysis. We have developed a novel approach for nuclear segmentation in histology sections, which addresses the problem of technical and biological variations by incorporating information from manually annotated reference patches with the local color space of the original image. Subsequently, the problem is formulated within a multi-reference graph cut with geodesic constraints. This approach has been validated on manually curated samples and then applied to a dataset of 440 whole mount tissue sections, originating from different laboratories, which are typically 40k-by-40k pixels or larger. Segmentation results, through a zoomable interface, and extracted morphometric data are available at: http://tcga.lbl.gov.

KW - H&E tissue section

KW - Nuclear segmentation

KW - Nuclear/Background classification

UR - http://www.scopus.com/inward/record.url?scp=84864856917&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84864856917&partnerID=8YFLogxK

U2 - 10.1109/ISBI.2012.6235683

DO - 10.1109/ISBI.2012.6235683

M3 - Conference contribution

AN - SCOPUS:84864856917

SN - 9781457718588

SP - 856

EP - 859

BT - Proceedings - International Symposium on Biomedical Imaging

ER -