Quantitative feature analysis in pathology
Principal Investigator
Name
Daniel Rubin
Institution
Stanford University
Position Title
Asst. Professor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-40
Initial CDAS Request Approval
Oct 30, 2013
Title
Quantitative feature analysis in pathology
Summary
Our goal is to develop and to identify quantitative image features in digital pathology images that stratify survival in the NLST patients. We will develop quantitative image features, extract them from the pathology images, and develop machine learning methods to predict survival from those features or stratify survival groups using those features. We will also look to develop quantitative image features correlations with the radiology images.
Aims
1. develop quantitative image features characterizing cancers on digital pathology images
2. build machine learning methods to predict survival or survival groups from the iamges features
3. evaluate our methods in a held-out dataset from NLST.