Abstract
Digital pathology imaging of tumor tissues, which captures histological details in high resolution, is fast becoming a routine clinical procedure. Recent developments in deep-learning methods have enabled the identification, characterization, and classification of individual cells from pathology images analysis at a large scale. This creates new opportunities to study the spatial patterns of and interactions among different types of cells. Reliable statistical approaches to modeling such spatial patterns and interactions can provide insight into tumor progression and shed light on the biological mechanisms of cancer. In this article, we consider the problem of modeling a pathology image with irregular locations of three different types of cells: lymphocyte, stromal, and tumor cells. We propose a novel Bayesian hierarchical model, which incorporates a hidden Potts model to project the irregularly distributed cells to a square lattice and a Markov random field prior model to identify regions in a heterogeneous pathology image. The model allows us to quantify the interactions between different types of cells, some of which are clinically meaningful. We use Markov chain Monte Carlo sampling techniques, combined with a double Metropolis-Hastings algorithm, in order to simulate samples approximately from a distribution with an intractable normalizing constant. The proposed model was applied to the pathology images of $205$ lung cancer patients from the National Lung Screening trial, and the results show that the interaction strength between tumor and stromal cells predicts patient prognosis (P = $0.005$). This statistical methodology provides a new perspective for understanding the role of cell-cell interactions in cancer progression.
Original language | English (US) |
---|---|
Pages (from-to) | 565-581 |
Number of pages | 17 |
Journal | Biostatistics (Oxford, England) |
Volume | 20 |
Issue number | 4 |
DOIs | |
State | Published - Oct 1 2019 |
Fingerprint
Keywords
- Double Metropolis–Hastings
- Hidden Potts model
- Lung cancer
- Markov random field
- Mixture model
- Pathology image
- Potts model
- Spatial point pattern
ASJC Scopus subject areas
- Statistics and Probability
- Statistics, Probability and Uncertainty
Cite this
A Bayesian hidden Potts mixture model for analyzing lung cancer pathology images. / Li, Qiwei; Wang, Xinlei; Liang, Faming; Yi, Faliu; Xie, Yang; Gazdar, Adi; Xiao, Guanghua.
In: Biostatistics (Oxford, England), Vol. 20, No. 4, 01.10.2019, p. 565-581.Research output: Contribution to journal › Article
}
TY - JOUR
T1 - A Bayesian hidden Potts mixture model for analyzing lung cancer pathology images
AU - Li, Qiwei
AU - Wang, Xinlei
AU - Liang, Faming
AU - Yi, Faliu
AU - Xie, Yang
AU - Gazdar, Adi
AU - Xiao, Guanghua
PY - 2019/10/1
Y1 - 2019/10/1
N2 - Digital pathology imaging of tumor tissues, which captures histological details in high resolution, is fast becoming a routine clinical procedure. Recent developments in deep-learning methods have enabled the identification, characterization, and classification of individual cells from pathology images analysis at a large scale. This creates new opportunities to study the spatial patterns of and interactions among different types of cells. Reliable statistical approaches to modeling such spatial patterns and interactions can provide insight into tumor progression and shed light on the biological mechanisms of cancer. In this article, we consider the problem of modeling a pathology image with irregular locations of three different types of cells: lymphocyte, stromal, and tumor cells. We propose a novel Bayesian hierarchical model, which incorporates a hidden Potts model to project the irregularly distributed cells to a square lattice and a Markov random field prior model to identify regions in a heterogeneous pathology image. The model allows us to quantify the interactions between different types of cells, some of which are clinically meaningful. We use Markov chain Monte Carlo sampling techniques, combined with a double Metropolis-Hastings algorithm, in order to simulate samples approximately from a distribution with an intractable normalizing constant. The proposed model was applied to the pathology images of $205$ lung cancer patients from the National Lung Screening trial, and the results show that the interaction strength between tumor and stromal cells predicts patient prognosis (P = $0.005$). This statistical methodology provides a new perspective for understanding the role of cell-cell interactions in cancer progression.
AB - Digital pathology imaging of tumor tissues, which captures histological details in high resolution, is fast becoming a routine clinical procedure. Recent developments in deep-learning methods have enabled the identification, characterization, and classification of individual cells from pathology images analysis at a large scale. This creates new opportunities to study the spatial patterns of and interactions among different types of cells. Reliable statistical approaches to modeling such spatial patterns and interactions can provide insight into tumor progression and shed light on the biological mechanisms of cancer. In this article, we consider the problem of modeling a pathology image with irregular locations of three different types of cells: lymphocyte, stromal, and tumor cells. We propose a novel Bayesian hierarchical model, which incorporates a hidden Potts model to project the irregularly distributed cells to a square lattice and a Markov random field prior model to identify regions in a heterogeneous pathology image. The model allows us to quantify the interactions between different types of cells, some of which are clinically meaningful. We use Markov chain Monte Carlo sampling techniques, combined with a double Metropolis-Hastings algorithm, in order to simulate samples approximately from a distribution with an intractable normalizing constant. The proposed model was applied to the pathology images of $205$ lung cancer patients from the National Lung Screening trial, and the results show that the interaction strength between tumor and stromal cells predicts patient prognosis (P = $0.005$). This statistical methodology provides a new perspective for understanding the role of cell-cell interactions in cancer progression.
KW - Double Metropolis–Hastings
KW - Hidden Potts model
KW - Lung cancer
KW - Markov random field
KW - Mixture model
KW - Pathology image
KW - Potts model
KW - Spatial point pattern
UR - http://www.scopus.com/inward/record.url?scp=85073562318&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85073562318&partnerID=8YFLogxK
U2 - 10.1093/biostatistics/kxy019
DO - 10.1093/biostatistics/kxy019
M3 - Article
C2 - 29788035
AN - SCOPUS:85073562318
VL - 20
SP - 565
EP - 581
JO - Biostatistics
JF - Biostatistics
SN - 1465-4644
IS - 4
ER -