Automated Feature Learning for Early Diagnosis of Lung Cancer using Low-Dose CT Scans

Principal Investigator

Name
Andrew Berlin

Degrees
Ph.D.

Institution
Draper

Position Title
Distinguished Member of Technical Staff

Email
aberlin@draper.com

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-417

Initial CDAS Request Approval
Jun 5, 2018

Title
Automated Feature Learning for Early Diagnosis of Lung Cancer using Low-Dose CT Scans

Summary
This project aims to develop deep learning methods to automatically learn informative and discriminatory features for Computer Aided Detection (CADe) of lung nodules and Computer Aided Diagnosis (CADx) of lung cancer.

Aims

- Develop deep learning based CADe models for lung nodule detection.
- Develop deep learning based CADx models for lung cancer diagnosis.
- Analyze and interpret automatically learned features for CADe and CADx.

Collaborators

Onur Ozdemir (Draper)
Rebecca L. Russell (Draper)