Development of deep learning frameworks based on pathology images for biomarker development

Principal Investigator

Name
Jennifer Beane

Degrees
Ph.D.

Institution
Trustees of Boston University

Position Title
Assistant Professor

Email
jbeane@bu.edu

About this CDAS Project

Study
PLCO (Learn more about this study)

Project ID
PLCOI-881

Initial CDAS Request Approval
Jan 5, 2022

Title
Development of deep learning frameworks based on pathology images for biomarker development

Summary
Our group is interested in developing deep learning frameworks to learn histologic patterns in lung tissue. We have developed different methods that can be applied to H&E whole slide images to predict lung cancer and lung cancer subtype. We would like to gain access to the PLCO pathology images to test the performance of our methods that were developed on the data from the NCI’s Clinical Proteomic Tumor Analysis consortium (CPTAC). As part of our work, we are also interested in understanding preinvasive lung lesions. Our overall goal is to look for lung cancer-associated histologic features in preinvasive lung lesions and assess whether or not they can enhance biomarker of preinvasive lesion progression.

Aims

Aim 1. Test the performance of patch-based deep learning frameworks to predict lung cancer and lung cancer subtypes on H&E whole slide images of lung tissue from the PLCO.

Aim 2. Test the performance whole slide and graph-based deep learning frameworks to predict lung cancer and lung cancer subtypes on H&E whole slide images of lung tissue from the PLCO.

Collaborators

Vijaya Kolachalama, Boston University School of Medicine