Skip to Main Content

An official website of the United States government

Automatic detection of pulmonary, cardiovascular and skeletal findings and their relation to all-cause mortality in NLST data

Principal Investigator
Name
Ivana Isgum
Degrees
PhD
Institution
Amsterdam University Medical Center
Position Title
Professor
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-111
Initial CDAS Request Approval
Jan 13, 2015
Title
Automatic detection of pulmonary, cardiovascular and skeletal findings and their relation to all-cause mortality in NLST data
Summary
It has been shown that participants in lung cancer screening trials are affected by fatal and nonfatal cardiovascular events (CVE) and other causes of death. We have developed algorithms for quantification of cardiovascular calcifications in chest CT and subsequently shown that CVE can be predicted based on these calcium scores. These algorithms have been developed using data from the NELSON trial where the data has been acquired using a single imaging protocol.
We propose to validate our previously developed algorithms for automatic prediction of all-cause mortality based on coronary and aortic calcium in lung cancer screening participants and when needed, further adjust the methodology to ensure robustness to image acquisition and imaged population.
Next, it has been shown that pulmonary and skeletal abnormalities independently predict mortality in the NELSON trial. Hence, we would like to identify association of cardiovascular calcifications, pulmonary and skeletal abnormalities with CVE, and disease specific and all-cause mortality in NLST.
Aims

(1) To validate performance of our previously developed algorithms for calcium scoring in the cardiovascular structures and investigate their progression, size, spatial and intensity characteristics.
(2) To perform automatic and visual analysis of cardiovascular, pulmonary and skeletal abnormalities.
(3) To relate these findings with CVE, and disease specific and all-cause mortality.

To enable case-cohort analysis, we would like to have an access to images of all subjects who died from the CT arm (N=1877) and a random sample of twice as many controls (N=3754). To analyze progression of cardiovascular calcifications, we would like to have access to the follow-up CT images of the included subjects.

Collaborators

Pim de Jong, MD PhD, University Medical Center Utrecht, Utrecht, The Netherlands
Bram van Ginneken, PhD, Radboud University Medical Center, Nijmegen, The Netherlands
Firdaus Mohamed Hoesein, MD PhD, University Medical Center Utrecht, Utrecht, The Netherlands Nikolas Lessmann, PhD student, University Medical Center Utrecht, Utrecht, The Netherlands
Bob de Vos, PhD student, University Medical Center Utrecht, Utrecht, The Netherlands
Jurica Sprem, PhD student, University Medical Center Utrecht, Utrecht, The Netherlands
Richard Takx, PhD student, University Medical Center Utrecht, Utrecht, The Netherlands
Colin Jacobs, PhD student, Radboud University Medical Center, Nijmegen, The Netherlands

Related Publications