Skip to Main Content

COVID-19 is an emerging, rapidly evolving situation.

What people with cancer should know: https://www.cancer.gov/coronavirus

Get the latest public health information from CDC: https://www.coronavirus.gov

Get the latest research information from NIH: https://covid19.nih.gov/

Principal Investigator
Name
Ryan Chamberlain
Degrees
PhD
Institution
Minnesota HealthSolutions
Position Title
Principal Research Scientist
Email
About this CDAS Project
Study
NLST (Learn more about this study)
Project ID
NLST-716
Initial CDAS Request Approval
Oct 8, 2020
Title
Automated Coronary Calcium Screening from Chest CT
Summary
Minnesota HealthSolutions (MHS) proposes a project to develop and validate a software product capable of automatically detecting and staging coronary artery calcium using low dose, ungated chest CT exams.  The proposed system will combine state-of-the-art machine learning methods and the clinical expertise of Brigham and Women’s Hospital into a system that integrates seamlessly into the Radiology workflow and standard patient care path to improve the treatment decisions for patients at risk for coronary artery disease.  The World Health Organization lists cardiovascular disease as the most common cause of death worldwide at 31% of all deaths.  Coronary artery calcium (CAC) scoring has become the most reliable predictor of future coronary artery disease (CAD) events in the asymptomatic population.  A successful completion of this project would provide a workflow-integrated tool capable of screening for coronary artery disease among the millions of ungated thoracic scans done annually in the United States, without additional radiation, cost, or patient burden.
Aims

Specific Aim 1: Implement and validate an automated coronary artery calcium detection algorithm in low dose chest CTs.  Fully automated CAC detection has historically been a difficult problem due to high false positive rates.  In this aim we will avoid the computationally expensive process of detecting individual calcium lesions and instead train a CNN to give a binary decision on two CAC score values: 1) > 100, and 2) > 300. 

Specific Aim 2: Implement and validate an automated coronary artery calcium localization algorithm in low dose chest CTs.  SA1 concentrates on the binary detection of CAC at two threshold levels: CAC > 100 and CAC > 300.  SA2 will expand the software to be able to identify individual calcium deposits, enabling automated quantitative scoring.

Collaborators

None