Automatic Coronary Calcium Scoring on both chest and cardiac CT using deep learning
Principal Investigator
Name
Yufeng Deng
Degrees
Ph.D.
Institution
Infervision US Inc.
Position Title
President
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-496
Initial CDAS Request Approval
Apr 9, 2019
Title
Automatic Coronary Calcium Scoring on both chest and cardiac CT using deep learning
Summary
Cardiovascular disease (CVD) is the global leading
cause of death. A strong risk factor for CVD events is the amount
of coronary artery calcium (CAC).
Current automatic calcium scoring methods mostly provide scores for only cardiac CT. To address this, we propose a new network based on Dense U-net to segment CAC pixels on both cardiac and chest CT. After segmenting lesion pixels, we use Agatston Score to calculate CAC scoring and evaluate the risk of CVD for patients. We believe that our method will enhance the efficiency and accuracy of radiologists.
Current automatic calcium scoring methods mostly provide scores for only cardiac CT. To address this, we propose a new network based on Dense U-net to segment CAC pixels on both cardiac and chest CT. After segmenting lesion pixels, we use Agatston Score to calculate CAC scoring and evaluate the risk of CVD for patients. We believe that our method will enhance the efficiency and accuracy of radiologists.
Aims
1: Develop an automatic method to compute coronary calcium scoring on the chest and cardiac CT
2: Get better results than previous work using NLST dataset.
Collaborators
NA