Skip to Main Content

An official website of the United States government

Government Funding Lapse

Because of a lapse in government funding, the information on this website may not be up to date, transactions submitted via the website may not be processed, and the agency may not be able to respond to inquiries until appropriations are enacted. The NIH Clinical Center (the research hospital of NIH) is open. For more details about its operating status, please visit  cc.nih.gov. Updates regarding government operating status and resumption of normal operations can be found at OPM.gov.

Principal Investigator
Name
Robert Rowley
Degrees
M.D.
Institution
Flow Health, Inc.
Position Title
Chief Medical Officer
Email
About this CDAS Project
Study
PLCO (Learn more about this study)
Project ID
PLCO-296
Initial CDAS Request Approval
Sep 5, 2017
Title
Detection and classification of abnormalities in Chest X-Rays
Summary
We propose using deep learning methods to build algorithms for classification of chest X-rays (CXRs) in the diseases/findings available in the PLCO dataset. More specifically we are planning to use deep convolutional neural networks (CNNs) and other deep learning architectures to help understand and identify normal and abnormal appearances of diseases in CXRs.
Aims

Aim 1: We are planning to build classifiers using CNNs to help classify CXRs. We believe that because PLCO dataset offers a large number of CXRs along with annotations the CNNs will be able to be trained and learn various abnormalities. More specifically we are planning to build:
1) a classifier for normal versus abnormal (cancer) CXRs
2) a classifier or classifiers for normal versus other diseases/conditions found in the PLCO dataset (e.g. nodules, emphysema, etc)
Aim 2: Using radiologists' input we are planning to annotate CXRs. Using trained classifiers as well as radiologists' ground truths the classifiers from Aim 1 will be able to improve and provide the location of the diseases/conditions.
Aim 3: We are planning to enhance the classifiers of Aim 1 and Aim 2 by incorporating metadata (i.e. patients' non-imaging information: age, gender, history, etc.). We believe that this additional data will provide insightful information and potential correlations.

Collaborators

Alex Meshkin - CEO
Dmitry Shingarev - Machine Learning Director