Computational tools for classification and histologic grading of lung cancer

Principal Investigator

Name
Ruijiang Li

Degrees
PhD

Institution
Stanford University

Position Title
Assistant Professor

Email
rli2@stanford.edu

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-169

Initial CDAS Request Approval
Oct 15, 2015

Title
Computational tools for classification and histologic grading of lung cancer

Summary
Since the new 2011 IASLC/ATS/ERS classification of lung adenocarcinoma was introduced, there have been a growing number of studies demonstrating the utility of comprehensive histological subtyping in identifying significant prognostic subsets of patients. However, a major obstacle to its implementation in routine clinical practice is the inter-observer variations and limited reproducibility, where a medium to low kappa value was reported even among expert lung cancer pathologists. In addition, there is no established histologic grading system for most lung cancers. In this project, we propose to develop and validate computational tools for automated, reliable classification and histologic grading of lung cancer based on H&E images in the NLST cohort.

Aims

1. Develop computational tools to automatically extract quantitative image features from tissue microarray data.
2. Develop methods to classify and grade cancers based on the extracted image features.
3. Evaluate the algorithm’s classification accuracy as well as the prognostic value by using cross validation.