Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks

Faliu Yi, Lin Yang, Shidan Wang, Lei Guo, Chenglong Huang, Yang Xie, Guanghua Xiao

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Background: Pathological angiogenesis has been identified in many malignancies as a potential prognostic factor and target for therapy. In most cases, angiogenic analysis is based on the measurement of microvessel density (MVD) detected by immunostaining of CD31 or CD34. However, most retrievable public data is generally composed of Hematoxylin and Eosin (H&E)-stained pathology images, for which is difficult to get the corresponding immunohistochemistry images. The role of microvessels in H&E stained images has not been widely studied due to their complexity and heterogeneity. Furthermore, identifying microvessels manually for study is a labor-intensive task for pathologists, with high inter- and intra-observer variation. Therefore, it is important to develop automated microvessel-detection algorithms in H&E stained pathology images for clinical association analysis. Results: In this paper, we propose a microvessel prediction method using fully convolutional neural networks. The feasibility of our proposed algorithm is demonstrated through experimental results on H&E stained images. Furthermore, the identified microvessel features were significantly associated with the patient clinical outcomes. Conclusions: This is the first study to develop an algorithm for automated microvessel detection in H&E stained pathology images.

Original languageEnglish (US)
Article number64
JournalBMC Bioinformatics
Volume19
Issue number1
DOIs
StatePublished - Feb 27 2018

Fingerprint

Pathology
Microvessels
Neural Networks
Neural networks
Prediction
Hematoxylin
Eosine Yellowish-(YS)
Observer
Immunohistochemistry
Personnel
Prognostic Factors
Angiogenesis
Pathologic Neovascularization
Clinical Pathology
Observer Variation
Therapy
Target
Experimental Results
Neoplasms

Keywords

  • Angiogenesis
  • Fully convolutional neural networks
  • H&E images
  • Microvessel
  • Pathology image

ASJC Scopus subject areas

  • Structural Biology
  • Biochemistry
  • Molecular Biology
  • Computer Science Applications
  • Applied Mathematics

Cite this

Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks. / Yi, Faliu; Yang, Lin; Wang, Shidan; Guo, Lei; Huang, Chenglong; Xie, Yang; Xiao, Guanghua.

In: BMC Bioinformatics, Vol. 19, No. 1, 64, 27.02.2018.

Research output: Contribution to journalArticle

Yi, Faliu ; Yang, Lin ; Wang, Shidan ; Guo, Lei ; Huang, Chenglong ; Xie, Yang ; Xiao, Guanghua. / Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks. In: BMC Bioinformatics. 2018 ; Vol. 19, No. 1.
@article{d97fb0725ed240a39141a47b46846b72,
title = "Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks",
abstract = "Background: Pathological angiogenesis has been identified in many malignancies as a potential prognostic factor and target for therapy. In most cases, angiogenic analysis is based on the measurement of microvessel density (MVD) detected by immunostaining of CD31 or CD34. However, most retrievable public data is generally composed of Hematoxylin and Eosin (H&E)-stained pathology images, for which is difficult to get the corresponding immunohistochemistry images. The role of microvessels in H&E stained images has not been widely studied due to their complexity and heterogeneity. Furthermore, identifying microvessels manually for study is a labor-intensive task for pathologists, with high inter- and intra-observer variation. Therefore, it is important to develop automated microvessel-detection algorithms in H&E stained pathology images for clinical association analysis. Results: In this paper, we propose a microvessel prediction method using fully convolutional neural networks. The feasibility of our proposed algorithm is demonstrated through experimental results on H&E stained images. Furthermore, the identified microvessel features were significantly associated with the patient clinical outcomes. Conclusions: This is the first study to develop an algorithm for automated microvessel detection in H&E stained pathology images.",
keywords = "Angiogenesis, Fully convolutional neural networks, H&E images, Microvessel, Pathology image",
author = "Faliu Yi and Lin Yang and Shidan Wang and Lei Guo and Chenglong Huang and Yang Xie and Guanghua Xiao",
year = "2018",
month = "2",
day = "27",
doi = "10.1186/s12859-018-2055-z",
language = "English (US)",
volume = "19",
journal = "BMC Bioinformatics",
issn = "1471-2105",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - Microvessel prediction in H&E Stained Pathology Images using fully convolutional neural networks

AU - Yi, Faliu

AU - Yang, Lin

AU - Wang, Shidan

AU - Guo, Lei

AU - Huang, Chenglong

AU - Xie, Yang

AU - Xiao, Guanghua

PY - 2018/2/27

Y1 - 2018/2/27

N2 - Background: Pathological angiogenesis has been identified in many malignancies as a potential prognostic factor and target for therapy. In most cases, angiogenic analysis is based on the measurement of microvessel density (MVD) detected by immunostaining of CD31 or CD34. However, most retrievable public data is generally composed of Hematoxylin and Eosin (H&E)-stained pathology images, for which is difficult to get the corresponding immunohistochemistry images. The role of microvessels in H&E stained images has not been widely studied due to their complexity and heterogeneity. Furthermore, identifying microvessels manually for study is a labor-intensive task for pathologists, with high inter- and intra-observer variation. Therefore, it is important to develop automated microvessel-detection algorithms in H&E stained pathology images for clinical association analysis. Results: In this paper, we propose a microvessel prediction method using fully convolutional neural networks. The feasibility of our proposed algorithm is demonstrated through experimental results on H&E stained images. Furthermore, the identified microvessel features were significantly associated with the patient clinical outcomes. Conclusions: This is the first study to develop an algorithm for automated microvessel detection in H&E stained pathology images.

AB - Background: Pathological angiogenesis has been identified in many malignancies as a potential prognostic factor and target for therapy. In most cases, angiogenic analysis is based on the measurement of microvessel density (MVD) detected by immunostaining of CD31 or CD34. However, most retrievable public data is generally composed of Hematoxylin and Eosin (H&E)-stained pathology images, for which is difficult to get the corresponding immunohistochemistry images. The role of microvessels in H&E stained images has not been widely studied due to their complexity and heterogeneity. Furthermore, identifying microvessels manually for study is a labor-intensive task for pathologists, with high inter- and intra-observer variation. Therefore, it is important to develop automated microvessel-detection algorithms in H&E stained pathology images for clinical association analysis. Results: In this paper, we propose a microvessel prediction method using fully convolutional neural networks. The feasibility of our proposed algorithm is demonstrated through experimental results on H&E stained images. Furthermore, the identified microvessel features were significantly associated with the patient clinical outcomes. Conclusions: This is the first study to develop an algorithm for automated microvessel detection in H&E stained pathology images.

KW - Angiogenesis

KW - Fully convolutional neural networks

KW - H&E images

KW - Microvessel

KW - Pathology image

UR - http://www.scopus.com/inward/record.url?scp=85042550844&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042550844&partnerID=8YFLogxK

U2 - 10.1186/s12859-018-2055-z

DO - 10.1186/s12859-018-2055-z

M3 - Article

VL - 19

JO - BMC Bioinformatics

JF - BMC Bioinformatics

SN - 1471-2105

IS - 1

M1 - 64

ER -