A bayesian mark interaction model for analysis of tumor pathology images

By Qiwei Li, Xinlei Wang, Faming Liang, Guanghua Xiao

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

With the advance of imaging technology, digital pathology imaging of tumor tissue slides is becoming a routine clinical procedure for cancer diagnosis. This process produces massive imaging data that capture histological details in high resolution. Recent developments in deep-learning methods have enabled us to identify and classify individual cells from digital pathology images at large scale. Reliable statistical approaches to model the spatial pattern of cells can provide new insight into tumor progression and shed light on the biological mechanisms of cancer. We consider the problem of modeling spatial correlations among three commonly seen cells observed in tumor pathology images. A novel geostatistical marking model with interpretable underlying parameters is proposed in a Bayesian framework. We use auxiliary variable MCMC algorithms to sample from the posterior distribution with an intractable normalizing constant. We demonstrate how this model-based analysis can lead to sharper inferences than ordinary exploratory analyses, by means of application to three benchmark datasets and a case study on the pathology images of 188 lung cancer patients. The case study shows that the spatial correlation between tumor and stromal cells predicts patient prognosis. This statistical methodology not only presents a new model for characterizing spatial correlations in a multitype spatial point pattern conditioning on the locations of the points, but also provides a new perspective for understanding the role of cell–cell interactions in cancer progression.

Original languageEnglish (US)
Pages (from-to)1708-1732
Number of pages25
JournalAnnals of Applied Statistics
Volume13
Issue number3
DOIs
StatePublished - Sep 2019

Fingerprint

Pathology
Tumors
Tumor
Spatial Correlation
Cancer
Cell
Progression
Interaction
Imaging techniques
Spatial Point Pattern
Imaging
MCMC Algorithm
Normalizing Constant
Digital Imaging
Multitype
Auxiliary Variables
Lung Cancer
Prognosis
Spatial Pattern
Posterior distribution

Keywords

  • Double Metropolis-Hastings
  • Markov random field
  • Multitype point pattern
  • Spatial correlation

ASJC Scopus subject areas

  • Statistics and Probability
  • Modeling and Simulation
  • Statistics, Probability and Uncertainty

Cite this

A bayesian mark interaction model for analysis of tumor pathology images. / Li, By Qiwei; Wang, Xinlei; Liang, Faming; Xiao, Guanghua.

In: Annals of Applied Statistics, Vol. 13, No. 3, 09.2019, p. 1708-1732.

Research output: Contribution to journalArticle

Li, By Qiwei ; Wang, Xinlei ; Liang, Faming ; Xiao, Guanghua. / A bayesian mark interaction model for analysis of tumor pathology images. In: Annals of Applied Statistics. 2019 ; Vol. 13, No. 3. pp. 1708-1732.
@article{a10600f7b8fe4bc3a5985c04bcd92170,
title = "A bayesian mark interaction model for analysis of tumor pathology images",
abstract = "With the advance of imaging technology, digital pathology imaging of tumor tissue slides is becoming a routine clinical procedure for cancer diagnosis. This process produces massive imaging data that capture histological details in high resolution. Recent developments in deep-learning methods have enabled us to identify and classify individual cells from digital pathology images at large scale. Reliable statistical approaches to model the spatial pattern of cells can provide new insight into tumor progression and shed light on the biological mechanisms of cancer. We consider the problem of modeling spatial correlations among three commonly seen cells observed in tumor pathology images. A novel geostatistical marking model with interpretable underlying parameters is proposed in a Bayesian framework. We use auxiliary variable MCMC algorithms to sample from the posterior distribution with an intractable normalizing constant. We demonstrate how this model-based analysis can lead to sharper inferences than ordinary exploratory analyses, by means of application to three benchmark datasets and a case study on the pathology images of 188 lung cancer patients. The case study shows that the spatial correlation between tumor and stromal cells predicts patient prognosis. This statistical methodology not only presents a new model for characterizing spatial correlations in a multitype spatial point pattern conditioning on the locations of the points, but also provides a new perspective for understanding the role of cell–cell interactions in cancer progression.",
keywords = "Double Metropolis-Hastings, Markov random field, Multitype point pattern, Spatial correlation",
author = "Li, {By Qiwei} and Xinlei Wang and Faming Liang and Guanghua Xiao",
year = "2019",
month = "9",
doi = "10.1214/19-AOAS1254",
language = "English (US)",
volume = "13",
pages = "1708--1732",
journal = "Annals of Applied Statistics",
issn = "1932-6157",
publisher = "Institute of Mathematical Statistics",
number = "3",

}

TY - JOUR

T1 - A bayesian mark interaction model for analysis of tumor pathology images

AU - Li, By Qiwei

AU - Wang, Xinlei

AU - Liang, Faming

AU - Xiao, Guanghua

PY - 2019/9

Y1 - 2019/9

N2 - With the advance of imaging technology, digital pathology imaging of tumor tissue slides is becoming a routine clinical procedure for cancer diagnosis. This process produces massive imaging data that capture histological details in high resolution. Recent developments in deep-learning methods have enabled us to identify and classify individual cells from digital pathology images at large scale. Reliable statistical approaches to model the spatial pattern of cells can provide new insight into tumor progression and shed light on the biological mechanisms of cancer. We consider the problem of modeling spatial correlations among three commonly seen cells observed in tumor pathology images. A novel geostatistical marking model with interpretable underlying parameters is proposed in a Bayesian framework. We use auxiliary variable MCMC algorithms to sample from the posterior distribution with an intractable normalizing constant. We demonstrate how this model-based analysis can lead to sharper inferences than ordinary exploratory analyses, by means of application to three benchmark datasets and a case study on the pathology images of 188 lung cancer patients. The case study shows that the spatial correlation between tumor and stromal cells predicts patient prognosis. This statistical methodology not only presents a new model for characterizing spatial correlations in a multitype spatial point pattern conditioning on the locations of the points, but also provides a new perspective for understanding the role of cell–cell interactions in cancer progression.

AB - With the advance of imaging technology, digital pathology imaging of tumor tissue slides is becoming a routine clinical procedure for cancer diagnosis. This process produces massive imaging data that capture histological details in high resolution. Recent developments in deep-learning methods have enabled us to identify and classify individual cells from digital pathology images at large scale. Reliable statistical approaches to model the spatial pattern of cells can provide new insight into tumor progression and shed light on the biological mechanisms of cancer. We consider the problem of modeling spatial correlations among three commonly seen cells observed in tumor pathology images. A novel geostatistical marking model with interpretable underlying parameters is proposed in a Bayesian framework. We use auxiliary variable MCMC algorithms to sample from the posterior distribution with an intractable normalizing constant. We demonstrate how this model-based analysis can lead to sharper inferences than ordinary exploratory analyses, by means of application to three benchmark datasets and a case study on the pathology images of 188 lung cancer patients. The case study shows that the spatial correlation between tumor and stromal cells predicts patient prognosis. This statistical methodology not only presents a new model for characterizing spatial correlations in a multitype spatial point pattern conditioning on the locations of the points, but also provides a new perspective for understanding the role of cell–cell interactions in cancer progression.

KW - Double Metropolis-Hastings

KW - Markov random field

KW - Multitype point pattern

KW - Spatial correlation

UR - http://www.scopus.com/inward/record.url?scp=85073788613&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85073788613&partnerID=8YFLogxK

U2 - 10.1214/19-AOAS1254

DO - 10.1214/19-AOAS1254

M3 - Article

AN - SCOPUS:85073788613

VL - 13

SP - 1708

EP - 1732

JO - Annals of Applied Statistics

JF - Annals of Applied Statistics

SN - 1932-6157

IS - 3

ER -