Skip to Main Content
An official website of the United States government
Upcoming Login Change: On or about August 24, CDAS will begin using the NIH Researcher Auth Service (RAS) to manage our existing login options. When logging into CDAS, you will be taken to RAS, where you can select your preferred login provider to access CDAS.

Prediction of abnormalities in Chest X-Rays with deep learning techniques

Principal Investigator

Name
Sasa Grbic

Degrees
Ph.D.

Institution
Siemens Medical Solutions

Position Title
Staff Scientist

Email
sasa.grbic@siemens-healthineers.com

About this CDAS Project

Study
NLST (Learn more about this study)

Project ID
NLST-397

Initial CDAS Request Approval
Apr 9, 2018

Title
Prediction of abnormalities in Chest X-Rays with deep learning techniques

Summary
Explore methods to predict abnormality in Chest X-Ray (collected within NLST data base) with novel deep learning technologies.

Aims

Aim 1: Building and testing existing deep neural networks to predict abnormality labels based on solely Chest X-Rays (validation on accuracy, specificity and sensitivity)
Aim 1: Building and testing existing deep neural networks to predict normal vs abnormal labels based on solely Chest X-Rays (validation on accuracy, specificity and sensitivity)

Collaborators

Sasa Grbic
Siqi Liu
Dominik Neumann
Sebastian Guendel
Bogdan Georgescu
Puneet Sharma