Bharath Ambale-Venkatesh1, Xiaoying Yang1, Colin O Wu1, Kiang Liu1, W Gregory Hundley1, Robyn McClelland1, Antoinette S Gomes1, Aaron R Folsom1, Steven Shea1, Eliseo Guallar1, David A Bluemke1, João A C Lima2. 1. From the Department of Radiology (B.A.-V.), Bloomberg School of Public Health (E.G.), and Department of Medicine, Cardiology and Radiology (J.A.C.L.), Johns Hopkins University, Baltimore, MD; George Washington University, DC (X.Y.); Office of Biostatistics, NHLBI, NIH, Bethesda, MD (C.O.W.); Department of Preventive Medicine, Northwestern University Medical School, Chicago, IL (K.L.); Department of Cardiology, Wake Forest University Health Sciences, Winston-Salem, NC (W.G.H.); Department of Biostatistics, University of Washington, Seattle (R.M.); Department of Radiology, UCLA School of Medicine, Los Angeles, CA (A.S.G.); Division of Epidemiology and Community Health, University of Minnesota, Minneapolis (A.R.F.); Departments of Medicine and Epidemiology, Columbia University, New York, NY (S.S.); and Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, MD (D.A.B.). 2. From the Department of Radiology (B.A.-V.), Bloomberg School of Public Health (E.G.), and Department of Medicine, Cardiology and Radiology (J.A.C.L.), Johns Hopkins University, Baltimore, MD; George Washington University, DC (X.Y.); Office of Biostatistics, NHLBI, NIH, Bethesda, MD (C.O.W.); Department of Preventive Medicine, Northwestern University Medical School, Chicago, IL (K.L.); Department of Cardiology, Wake Forest University Health Sciences, Winston-Salem, NC (W.G.H.); Department of Biostatistics, University of Washington, Seattle (R.M.); Department of Radiology, UCLA School of Medicine, Los Angeles, CA (A.S.G.); Division of Epidemiology and Community Health, University of Minnesota, Minneapolis (A.R.F.); Departments of Medicine and Epidemiology, Columbia University, New York, NY (S.S.); and Radiology and Imaging Sciences, NIH Clinical Center, Bethesda, MD (D.A.B.). jlima@jhmi.edu.
Abstract
RATIONALE: Machine learning may be useful to characterize cardiovascular risk, predict outcomes, and identify biomarkers in population studies. OBJECTIVE: To test the ability of random survival forests, a machine learning technique, to predict 6 cardiovascular outcomes in comparison to standard cardiovascular risk scores. METHODS AND RESULTS: We included participants from the MESA (Multi-Ethnic Study of Atherosclerosis). Baseline measurements were used to predict cardiovascular outcomes over 12 years of follow-up. MESA was designed to study progression of subclinical disease to cardiovascular events where participants were initially free of cardiovascular disease. All 6814 participants from MESA, aged 45 to 84 years, from 4 ethnicities, and 6 centers across the United States were included. Seven-hundred thirty-five variables from imaging and noninvasive tests, questionnaires, and biomarker panels were obtained. We used the random survival forests technique to identify the top-20 predictors of each outcome. Imaging, electrocardiography, and serum biomarkers featured heavily on the top-20 lists as opposed to traditional cardiovascular risk factors. Age was the most important predictor for all-cause mortality. Fasting glucose levels and carotid ultrasonography measures were important predictors of stroke. Coronary Artery Calcium score was the most important predictor of coronary heart disease and all atherosclerotic cardiovascular disease combined outcomes. Left ventricular structure and function and cardiac troponin-T were among the top predictors for incident heart failure. Creatinine, age, and ankle-brachial index were among the top predictors of atrial fibrillation. TNF-α (tissue necrosis factor-α) and IL (interleukin)-2 soluble receptors and NT-proBNP (N-Terminal Pro-B-Type Natriuretic Peptide) levels were important across all outcomes. The random survival forests technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 10%-25%). CONCLUSIONS:Machine learning in conjunction with deep phenotyping improves prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. These methods may lead to greater insights on subclinical disease markers without apriori assumptions of causality. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00005487.
RCT Entities:
RATIONALE: Machine learning may be useful to characterize cardiovascular risk, predict outcomes, and identify biomarkers in population studies. OBJECTIVE: To test the ability of random survival forests, a machine learning technique, to predict 6 cardiovascular outcomes in comparison to standard cardiovascular risk scores. METHODS AND RESULTS: We included participants from the MESA (Multi-Ethnic Study of Atherosclerosis). Baseline measurements were used to predict cardiovascular outcomes over 12 years of follow-up. MESA was designed to study progression of subclinical disease to cardiovascular events where participants were initially free of cardiovascular disease. All 6814 participants from MESA, aged 45 to 84 years, from 4 ethnicities, and 6 centers across the United States were included. Seven-hundred thirty-five variables from imaging and noninvasive tests, questionnaires, and biomarker panels were obtained. We used the random survival forests technique to identify the top-20 predictors of each outcome. Imaging, electrocardiography, and serum biomarkers featured heavily on the top-20 lists as opposed to traditional cardiovascular risk factors. Age was the most important predictor for all-cause mortality. Fasting glucose levels and carotid ultrasonography measures were important predictors of stroke. Coronary Artery Calcium score was the most important predictor of coronary heart disease and all atherosclerotic cardiovascular disease combined outcomes. Left ventricular structure and function and cardiac troponin-T were among the top predictors for incident heart failure. Creatinine, age, and ankle-brachial index were among the top predictors of atrial fibrillation. TNF-α (tissue necrosis factor-α) and IL (interleukin)-2 soluble receptors and NT-proBNP (N-Terminal Pro-B-Type Natriuretic Peptide) levels were important across all outcomes. The random survival forests technique performed better than established risk scores with increased prediction accuracy (decreased Brier score by 10%-25%). CONCLUSIONS: Machine learning in conjunction with deep phenotyping improves prediction accuracy in cardiovascular event prediction in an initially asymptomatic population. These methods may lead to greater insights on subclinical disease markers without apriori assumptions of causality. CLINICAL TRIAL REGISTRATION: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00005487.
Authors: Ryo Inuzuka; Gerhard-Paul Diller; Francesco Borgia; Leah Benson; Edgar L W Tay; Rafael Alonso-Gonzalez; Margarida Silva; Menelaos Charalambides; Lorna Swan; Konstantinos Dimopoulos; Michael A Gatzoulis Journal: Circulation Date: 2011-12-06 Impact factor: 29.690
Authors: Harjit Chahal; David A Bluemke; Colin O Wu; Robyn McClelland; Kiang Liu; Steven J Shea; Gregory Burke; Pelbreton Balfour; David Herrington; PeiBei Shi; Wendy Post; Jean Olson; Karol E Watson; Aaron R Folsom; Joao A C Lima Journal: Heart Date: 2014-11-07 Impact factor: 5.994
Authors: Ralph B D'Agostino; Ramachandran S Vasan; Michael J Pencina; Philip A Wolf; Mark Cobain; Joseph M Massaro; William B Kannel Journal: Circulation Date: 2008-01-22 Impact factor: 29.690
Authors: Joshua D Bundy; Susan R Heckbert; Lin Y Chen; Donald M Lloyd-Jones; Philip Greenland Journal: Am J Cardiol Date: 2019-10-10 Impact factor: 2.778
Authors: Dustin Hillerson; Thomas Wool; Gbolahan O Ogunbayo; Vincent L Sorrell; Steve W Leung Journal: AJR Am J Roentgenol Date: 2020-04-29 Impact factor: 3.959
Authors: Frederic Commandeur; Piotr J Slomka; Markus Goeller; Xi Chen; Sebastien Cadet; Aryabod Razipour; Priscilla McElhinney; Heidi Gransar; Stephanie Cantu; Robert J H Miller; Alan Rozanski; Stephan Achenbach; Balaji K Tamarappoo; Daniel S Berman; Damini Dey Journal: Cardiovasc Res Date: 2020-12-01 Impact factor: 10.787
Authors: Ankush Jamthikar; Deep Gupta; Narendra N Khanna; Tadashi Araki; Luca Saba; Andrew Nicolaides; Aditya Sharma; Tomaz Omerzu; Harman S Suri; Ajay Gupta; Sophie Mavrogeni; Monika Turk; John R Laird; Athanasios Protogerou; Petros P Sfikakis; George D Kitas; Vijay Viswanathan; Gyan Pareek; Martin Miner; Jasjit S Suri Journal: Curr Atheroscler Rep Date: 2019-05-01 Impact factor: 5.113
Authors: Ankush D Jamthikar; Deep Gupta; Laura E Mantella; Luca Saba; John R Laird; Amer M Johri; Jasjit S Suri Journal: Int J Cardiovasc Imaging Date: 2020-11-12 Impact factor: 2.357
Authors: Philip Greenland; Michael J Blaha; Matthew J Budoff; Raimund Erbel; Karol E Watson Journal: J Am Coll Cardiol Date: 2018-07-24 Impact factor: 24.094