Digital Assessment of Stained Breast Tissue Images for Comprehensive Tumor and Microenvironment Analysis

Shachi Mittal, Catalin Stoean, Andre Kajdacsy-Balla, Rohit Bhargava

Research output: Contribution to journalArticle

Abstract

Current histopathological diagnosis involves human expert interpretation of stained images for diagnosis. This process is prone to inter-observer variability, often leading to low concordance rates amongst pathologists across many types of tissues. Further, since structural features are mostly just defined for epithelial alterations during tumor progression, the use of associated stromal changes is limited. Here we sought to examine whether digital analysis of commonly used hematoxylin and eosin-stained images could provide precise and quantitative metrics of disease from both epithelial and stromal cells. We developed a convolutional neural network approach to identify epithelial breast cells from their microenvironment. Second, we analyzed the microenvironment to further observe different constituent cells using unsupervised clustering. Finally, we categorized breast cancer by the combined effects of stromal and epithelial inertia. Together, the work provides insight and evidence of cancer association for interpretable features from deep learning methods that provide new opportunities for comprehensive analysis of standard pathology images.

Original languageEnglish (US)
Article number246
JournalFrontiers in Bioengineering and Biotechnology
Volume7
DOIs
StatePublished - Oct 1 2019

Fingerprint

Tumor Microenvironment
Tumors
Breast
Epithelial Cells
Tissue
Cellular Microenvironment
Observer Variation
Pathology
Hematoxylin
Eosine Yellowish-(YS)
Stromal Cells
Cluster Analysis
Neoplasms
Learning
Breast Neoplasms
Neural networks
Pathologists
Deep learning

Keywords

  • breast cancer
  • deep learning
  • ductal carcinoma in-situ
  • hyperplasia and clustering
  • microenvironment

ASJC Scopus subject areas

  • Biotechnology
  • Bioengineering
  • Histology
  • Biomedical Engineering

Cite this

Digital Assessment of Stained Breast Tissue Images for Comprehensive Tumor and Microenvironment Analysis. / Mittal, Shachi; Stoean, Catalin; Kajdacsy-Balla, Andre; Bhargava, Rohit.

In: Frontiers in Bioengineering and Biotechnology, Vol. 7, 246, 01.10.2019.

Research output: Contribution to journalArticle

@article{aac87777cfc34784b11caccaa2afc6a7,
title = "Digital Assessment of Stained Breast Tissue Images for Comprehensive Tumor and Microenvironment Analysis",
abstract = "Current histopathological diagnosis involves human expert interpretation of stained images for diagnosis. This process is prone to inter-observer variability, often leading to low concordance rates amongst pathologists across many types of tissues. Further, since structural features are mostly just defined for epithelial alterations during tumor progression, the use of associated stromal changes is limited. Here we sought to examine whether digital analysis of commonly used hematoxylin and eosin-stained images could provide precise and quantitative metrics of disease from both epithelial and stromal cells. We developed a convolutional neural network approach to identify epithelial breast cells from their microenvironment. Second, we analyzed the microenvironment to further observe different constituent cells using unsupervised clustering. Finally, we categorized breast cancer by the combined effects of stromal and epithelial inertia. Together, the work provides insight and evidence of cancer association for interpretable features from deep learning methods that provide new opportunities for comprehensive analysis of standard pathology images.",
keywords = "breast cancer, deep learning, ductal carcinoma in-situ, hyperplasia and clustering, microenvironment",
author = "Shachi Mittal and Catalin Stoean and Andre Kajdacsy-Balla and Rohit Bhargava",
year = "2019",
month = "10",
day = "1",
doi = "10.3389/fbioe.2019.00246",
language = "English (US)",
volume = "7",
journal = "Frontiers in Bioengineering and Biotechnology",
issn = "2296-4185",
publisher = "Frontiers Media S. A.",

}

TY - JOUR

T1 - Digital Assessment of Stained Breast Tissue Images for Comprehensive Tumor and Microenvironment Analysis

AU - Mittal, Shachi

AU - Stoean, Catalin

AU - Kajdacsy-Balla, Andre

AU - Bhargava, Rohit

PY - 2019/10/1

Y1 - 2019/10/1

N2 - Current histopathological diagnosis involves human expert interpretation of stained images for diagnosis. This process is prone to inter-observer variability, often leading to low concordance rates amongst pathologists across many types of tissues. Further, since structural features are mostly just defined for epithelial alterations during tumor progression, the use of associated stromal changes is limited. Here we sought to examine whether digital analysis of commonly used hematoxylin and eosin-stained images could provide precise and quantitative metrics of disease from both epithelial and stromal cells. We developed a convolutional neural network approach to identify epithelial breast cells from their microenvironment. Second, we analyzed the microenvironment to further observe different constituent cells using unsupervised clustering. Finally, we categorized breast cancer by the combined effects of stromal and epithelial inertia. Together, the work provides insight and evidence of cancer association for interpretable features from deep learning methods that provide new opportunities for comprehensive analysis of standard pathology images.

AB - Current histopathological diagnosis involves human expert interpretation of stained images for diagnosis. This process is prone to inter-observer variability, often leading to low concordance rates amongst pathologists across many types of tissues. Further, since structural features are mostly just defined for epithelial alterations during tumor progression, the use of associated stromal changes is limited. Here we sought to examine whether digital analysis of commonly used hematoxylin and eosin-stained images could provide precise and quantitative metrics of disease from both epithelial and stromal cells. We developed a convolutional neural network approach to identify epithelial breast cells from their microenvironment. Second, we analyzed the microenvironment to further observe different constituent cells using unsupervised clustering. Finally, we categorized breast cancer by the combined effects of stromal and epithelial inertia. Together, the work provides insight and evidence of cancer association for interpretable features from deep learning methods that provide new opportunities for comprehensive analysis of standard pathology images.

KW - breast cancer

KW - deep learning

KW - ductal carcinoma in-situ

KW - hyperplasia and clustering

KW - microenvironment

UR - http://www.scopus.com/inward/record.url?scp=85073671021&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073671021&partnerID=8YFLogxK

U2 - 10.3389/fbioe.2019.00246

DO - 10.3389/fbioe.2019.00246

M3 - Article

AN - SCOPUS:85073671021

VL - 7

JO - Frontiers in Bioengineering and Biotechnology

JF - Frontiers in Bioengineering and Biotechnology

SN - 2296-4185

M1 - 246

ER -