Skip to Main Content

An official website of the United States government

Principal Investigator
Michael Lu
M.D., M.P.H.
Massachusetts General Hospital
Position Title
Director of Research and Assistant Professor of Radiology
About this CDAS Project
NLST (Learn more about this study)
Project ID
Initial CDAS Request Approval
Dec 29, 2017
Machine learning to predict risk from chest radiographs and clinical factors
Heart disease and cancer are the two most common causes of death in the developed world. NLST participants underwent screening chest radiography; analysis of these chest radiographs may be useful to predict incident morbidity and mortality. The goal of this proposal is to use machine learning to predict the risk of cancer, heart disease, and death from chest radiographs and other clinical information available in the NLST. Whether this approach is additive to existing risk prediction models will be assessed via questionnaire data about patient risk factors and medical history.

1) To predict cancer, heart disease, and death from chest radiographs using machine learning.
2) To compare the performance of this approach to existing risk prediction tools.


Parastou Eslami (Harvard)
Borek Foldyna (Harvard)
Udo Hoffmann (Harvard)
Alexander Ivanov (Harvard)
Chintan Parmar (Harvard)
Roman Zeleznik (Harvard)

Related Publications