|
Activity Number:
|
464
- Novel Approaches for Complex Biomedical Data
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
Section on Statistics in Genomics and Genetics
|
|
Abstract #313480
|
|
|
Title:
|
Multitype Replicated Point Pattern Analysis of Spatial Tumor-Immune Cell Interactions by Using Multivariate Log-Gaussian Cox Processes
|
|
Author(s):
|
Jooyoung Lee* and Molin Wang
|
|
Companies:
|
Harvard T.H. Chan School of Public Health and Harvard T.H. Chan School of Public Health
|
|
Keywords:
|
spatial point process;
multitype replicated point process;
tumor-immune microenvironment
|
|
Abstract:
|
Identifying the characteristics of tumor microenvironment is important to better understand cancer progression and to help treatment decision making. Recent advancement in digital pathology has enabled us to obtain not only the densities of different types of cells, but also histopathological image data from patient tissue samples. The image data then are converted into the location of each cell for each patient. Beyond the cell densities, the image data allow us to investigate the spatial pattern of each cell and the spatial association between tumor and immune cells, which may provide evidence for responses to immunotherapy. In multiplexed immunofluorescence (mIF) imaging data, images from multiple cores are available. We propose a model to quantify spatial association for multitype point processes when replicated data are available using multivariate log-gaussian Cox processes. We consider spatial pattern variation between multiple cores. We apply this method to the mIF imaging data in the Nurses' Health Study cohort.
|
Authors who are presenting talks have a * after their name.
