Coronary artery disease remains the leading cause of death worldwide, and more than half of the individuals suffering acute myocardial infarction have no prior symptoms. Coronary calcium scoring with non-contrast CT is increasingly used for cardiovascular risk stratification in the asymptomatic population. Epicardial adipose tissue (EAT), a local visceral fat depot surrounding the heart, has been recently reported as a significant imaging biomarker for cardiovascular risk stratification. Despite being routinely imaged noninvasively by non-contrast cardiac CT for coronary calcium scoring, EAT features are currently not measured or reported, primarily due to the absence of robust, automated quantification methods. We propose to employ novel image processing algorithms to achieve fully automated, robust quantification of EAT features from cardiac CT, and to apply machine learning methods to efficiently combine patient clinical data, coronary calcium and EAT features into a new integrated risk score to predict future cardiac events. We will evaluate this risk score in two existing, prospective registries of asymptomatic patients with available coronary calcium scans, clinical data, and followup for cardiac events (myocardial infarction, cardiac death, late revascularization). These rich resources are: the Early Identification of Subclinical Atherosclerosis Using Non-Invasive Imaging Research [EISNER] registry (2614 patients) and the St. Francis Heart Study (4613 patients). We propose three specific aims: 1) To develop and evaluate the accuracy and reproducibility of new computational algorithms for automated quantification of EAT measures; 2) To evaluate the prognostic value of automatically-quantified EAT features, relative to atherosclerotic plaque burden and measures of obesity, for the prediction of future cardiac events in asymptomatic patients in the EISNER registry and the St. Francis Heart Study; 3) To derive and evaluate a new, integrated risk score?combining standard risk factors, coronary calcium score and EAT measures using machine learning?for the prediction of future cardiac events in the EISNER registry, and further, to additionally evaluate this risk score externally in the St. Francis Heart Study patient cohort. This work in this proposal will provide a novel, personalized paradigm that will objectively and accurately identify asymptomatic patients undergoing coronary calcium scanning who are at increased risk for future cardiac events, without any additional imaging or risk to the patient.

Public Health Relevance

(lay language) CT scanning of the heart for coronary calcium scoring is a simple, widely-used, low-cost test with low radiation dose. Using these scans, the researchers propose to establish automated analysis of fat which surrounds the heart, as well as novel computerized scores that identify the patients who are at highest risk of suffering a heart attack. This new research will allow doctors to better identify patients for whom appropriate treatment could be prescribed, to avoid a heart attack or sudden cardiac death.

Agency
National Institute of Health (NIH)
Institute
National Heart, Lung, and Blood Institute (NHLBI)
Type
Research Project (R01)
Project #
5R01HL133616-02
Application #
9310380
Study Section
Cancer, Heart, and Sleep Epidemiology A Study Section (CHSA)
Program Officer
Danthi, Narasimhan
Project Start
2016-07-06
Project End
2020-04-30
Budget Start
2017-05-01
Budget End
2018-04-30
Support Year
2
Fiscal Year
2017
Total Cost
Indirect Cost
Name
Cedars-Sinai Medical Center
Department
Type
DUNS #
075307785
City
Los Angeles
State
CA
Country
United States
Zip Code
90048
Dey, Damini; Gaur, Sara; Ovrehus, Kristian A et al. (2018) Integrated prediction of lesion-specific ischaemia from quantitative coronary CT angiography using machine learning: a multicentre study. Eur Radiol 28:2655-2664
Goeller, Markus; Achenbach, Stephan; Marwan, Mohamed et al. (2018) Epicardial adipose tissue density and volume are related to subclinical atherosclerosis, inflammation and major adverse cardiac events in asymptomatic subjects. J Cardiovasc Comput Tomogr 12:67-73
Commandeur, Frederic; Goeller, Markus; Betancur, Julian et al. (2018) Deep Learning for Quantification of Epicardial and Thoracic Adipose Tissue From Non-Contrast CT. IEEE Trans Med Imaging 37:1835-1846
Nerlekar, Nitesh; Baey, Yi-Wei; Brown, Adam J et al. (2018) Poor Correlation, Reproducibility, and Agreement Between Volumetric Versus Linear Epicardial Adipose Tissue Measurement: A 3D Computed Tomography Versus 2D Echocardiography Comparison. JACC Cardiovasc Imaging 11:1035-1036
Dey, Damini; Commandeur, Frederic (2017) Radiomics to Identify High-Risk Atherosclerotic Plaque From Computed Tomography: The Power of Quantification. Circ Cardiovasc Imaging 10:
Hell, Michaela M; Motwani, Manish; Otaki, Yuka et al. (2017) Quantitative global plaque characteristics from coronary computed tomography angiography for the prediction of future cardiac mortality during long-term follow-up. Eur Heart J Cardiovasc Imaging 18:1331-1339
Slomka, Piotr J; Dey, Damini; Sitek, Arkadiusz et al. (2017) Cardiac imaging: working towards fully-automated machine analysis & interpretation. Expert Rev Med Devices 14:197-212