Julian Betancur1, Yuka Otaki1, Manish Motwani1, Mathews B Fish2, Mark Lemley2, Damini Dey1, Heidi Gransar1, Balaji Tamarappoo1, Guido Germano1, Tali Sharir3, Daniel S Berman1, Piotr J Slomka4. 1. Departments of Imaging, Medicine, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California. 2. Oregon Heart and Vascular Institute, Sacred Heart Medical Center, Springfield, Oregon. 3. Department of Nuclear Cardiology, Assuta Medical Centers, Tel Aviv, Israel. 4. Departments of Imaging, Medicine, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California. Electronic address: Piotr.Slomka@cshs.org.
Abstract
OBJECTIVES: This study evaluated the added predictive value of combining clinical information and myocardial perfusion single-photon emission computed tomography (SPECT) imaging (MPI) data using machine learning (ML) to predict major adverse cardiac events (MACE). BACKGROUND: Traditionally, prognostication by MPI has relied on visual or quantitative analysis of images without objective consideration of the clinical data. ML permits a large number of variables to be considered in combination and at a level of complexity beyond the human clinical reader. METHODS: A total of 2,619 consecutive patients (48% men; 62 ± 13 years of age) who underwent exercise (38%) or pharmacological stress (62%) with high-speed SPECT MPI were monitored for MACE. Twenty-eight clinical variables, 17 stress test variables, and 25 imaging variables (including total perfusion deficit [TPD]) were recorded. Areas under the receiver-operating characteristic curve (AUC) for MACE prediction were compared among: 1) ML with all available data (ML-combined); 2) ML with only imaging data (ML-imaging); 3) 5-point scale visual diagnosis (physician [MD] diagnosis); and 4) automated quantitative imaging analysis (stress TPD and ischemic TPD). ML involved automated variable selection by information gain ranking, model building with a boosted ensemble algorithm, and 10-fold stratified cross validation. RESULTS: During follow-up (3.2 ± 0.6 years), 239 patients (9.1%) had MACE. MACE prediction was significantly higher for ML-combined than ML-imaging (AUC: 0.81 vs. 0.78; p < 0.01). ML-combined also had higher predictive accuracy compared with MD diagnosis, automated stress TPD, and automated ischemic TPD (AUC: 0.81 vs. 0.65 vs. 0.73 vs. 0.71, respectively; p < 0.01 for all). Risk reclassification for ML-combined compared with visual MD diagnosis was 26% (p < 0.001). CONCLUSIONS: ML combined with both clinical and imaging data variables was found to have high predictive accuracy for 3-year risk of MACE and was superior to existing visual or automated perfusion assessments. ML could allow integration of clinical and imaging data for personalized MACE risk computations in patients undergoing SPECT MPI.
OBJECTIVES: This study evaluated the added predictive value of combining clinical information and myocardial perfusion single-photon emission computed tomography (SPECT) imaging (MPI) data using machine learning (ML) to predict major adverse cardiac events (MACE). BACKGROUND: Traditionally, prognostication by MPI has relied on visual or quantitative analysis of images without objective consideration of the clinical data. ML permits a large number of variables to be considered in combination and at a level of complexity beyond the human clinical reader. METHODS: A total of 2,619 consecutive patients (48% men; 62 ± 13 years of age) who underwent exercise (38%) or pharmacological stress (62%) with high-speed SPECT MPI were monitored for MACE. Twenty-eight clinical variables, 17 stress test variables, and 25 imaging variables (including total perfusion deficit [TPD]) were recorded. Areas under the receiver-operating characteristic curve (AUC) for MACE prediction were compared among: 1) ML with all available data (ML-combined); 2) ML with only imaging data (ML-imaging); 3) 5-point scale visual diagnosis (physician [MD] diagnosis); and 4) automated quantitative imaging analysis (stress TPD and ischemic TPD). ML involved automated variable selection by information gain ranking, model building with a boosted ensemble algorithm, and 10-fold stratified cross validation. RESULTS: During follow-up (3.2 ± 0.6 years), 239 patients (9.1%) had MACE. MACE prediction was significantly higher for ML-combined than ML-imaging (AUC: 0.81 vs. 0.78; p < 0.01). ML-combined also had higher predictive accuracy compared with MD diagnosis, automated stress TPD, and automated ischemic TPD (AUC: 0.81 vs. 0.65 vs. 0.73 vs. 0.71, respectively; p < 0.01 for all). Risk reclassification for ML-combined compared with visual MD diagnosis was 26% (p < 0.001). CONCLUSIONS: ML combined with both clinical and imaging data variables was found to have high predictive accuracy for 3-year risk of MACE and was superior to existing visual or automated perfusion assessments. ML could allow integration of clinical and imaging data for personalized MACE risk computations in patients undergoing SPECT MPI.
Authors: T Sharir; G Germano; X Kang; H C Lewin; R Miranda; I Cohen; R D Agafitei; J D Friedman; D S Berman Journal: J Nucl Med Date: 2001-06 Impact factor: 10.057
Authors: Ryo Nakazato; Balaji K Tamarappoo; Xingping Kang; Arik Wolak; Faith Kite; Sean W Hayes; Louise E J Thomson; John D Friedman; Daniel S Berman; Piotr J Slomka Journal: J Nucl Med Date: 2010-10-18 Impact factor: 10.057
Authors: Alessia Gimelli; Giuseppe Rossi; Patrizia Landi; Paolo Marzullo; Giorgio Iervasi; Antonio L'abbate; Daniele Rovai Journal: J Nucl Med Date: 2009-03-16 Impact factor: 10.057
Authors: Leslee J Shaw; Daniel S Berman; David J Maron; G B John Mancini; Sean W Hayes; Pamela M Hartigan; William S Weintraub; Robert A O'Rourke; Marcin Dada; John A Spertus; Bernard R Chaitman; John Friedman; Piotr Slomka; Gary V Heller; Guido Germano; Gilbert Gosselin; Peter Berger; William J Kostuk; Ronald G Schwartz; Merill Knudtson; Emir Veledar; Eric R Bates; Benjamin McCallister; Koon K Teo; William E Boden Journal: Circulation Date: 2008-02-11 Impact factor: 29.690
Authors: Julian Betancur; Lien-Hsin Hu; Frederic Commandeur; Tali Sharir; Andrew J Einstein; Mathews B Fish; Terrence D Ruddy; Philipp A Kaufmann; Albert J Sinusas; Edward J Miller; Timothy M Bateman; Sharmila Dorbala; Marcelo Di Carli; Guido Germano; Yuka Otaki; Joanna X Liang; Balaji K Tamarappoo; Damini Dey; Daniel S Berman; Piotr J Slomka Journal: J Nucl Med Date: 2018-09-27 Impact factor: 10.057
Authors: Yuka Otaki; Julian Betancur; Tali Sharir; Lien-Hsin Hu; Heidi Gransar; Joanna X Liang; Peyman N Azadani; Andrew J Einstein; Mathews B Fish; Terrence D Ruddy; Philipp A Kaufmann; Albert J Sinusas; Edward J Miller; Timothy M Bateman; Sharmila Dorbala; Marcelo Di Carli; Balaji K Tamarappoo; Guido Germano; Damini Dey; Daniel S Berman; Piotr J Slomka Journal: JACC Cardiovasc Imaging Date: 2019-06-12
Authors: Andrew Lin; Márton Kolossváry; Manish Motwani; Ivana Išgum; Pál Maurovich-Horvat; Piotr J Slomka; Damini Dey Journal: Radiol Cardiothorac Imaging Date: 2021-02-25
Authors: Julian Betancur; Frederic Commandeur; Mahsaw Motlagh; Tali Sharir; Andrew J Einstein; Sabahat Bokhari; Mathews B Fish; Terrence D Ruddy; Philipp Kaufmann; Albert J Sinusas; Edward J Miller; Timothy M Bateman; Sharmila Dorbala; Marcelo Di Carli; Guido Germano; Yuka Otaki; Balaji K Tamarappoo; Damini Dey; Daniel S Berman; Piotr J Slomka Journal: JACC Cardiovasc Imaging Date: 2018-03-14