Bayesian statistical approach to model cell-cell interaction for tumor pathology images analysis
Principal Investigator
Name
Alexandra Pershakova
Degrees
MSc
Institution
Politecnico di Milano
Position Title
Master Student
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-462
Initial CDAS Request Approval
Jan 2, 2019
Title
Bayesian statistical approach to model cell-cell interaction for tumor pathology images analysis
Summary
Studies of spatial patterns and interactions among different types of cells can provide clinicians with valuable insights into tissue disease progression, and can help to understand the underlying biological mechanisms. We aim to study in details the model for lymphocyte, stromal and tumor cells interaction proposed elsewhere (Li, Q. et al. 2018). We aim to show that model-based image analysis can be a useful technique for medical specialists. We turn to Bayesian approach and tools that provide solid foundation and variety of computational methods for development of complex models based on prior knowledge.
Aims
1. To build the model for mark interaction for describing the dynamics of lymphocyte, stromal and tumor cells.
2. Given the NLST Pathology Images, to consider the choice of parameters that would quantify the attraction and repulsion dynamics of cells.
3. Explore the possibility of using the model of the dynamics of lymphocyte, stromal and tumor cells for prognostic analysis.
Collaborators
Diana Isaeva, MSc Student at Politecnico di Milano