Author + information
- Received November 2, 2017
- Revision received January 15, 2018
- Accepted January 16, 2018
- Published online November 5, 2018.
- Julian Betancur, PhDa,
- Frederic Commandeur, PhDa,
- Mahsaw Motlagh, BAa,
- Tali Sharir, MDb,c,
- Andrew J. Einstein, MD, PhDd,e,
- Sabahat Bokhari, MDd,
- Mathews B. Fish, MDf,
- Terrence D. Ruddy, MDg,
- Philipp Kaufmann, MDh,
- Albert J. Sinusas, MDi,
- Edward J. Miller, MD, PhDi,
- Timothy M. Bateman, MDj,
- Sharmila Dorbala, MD, MPHk,
- Marcelo Di Carli, MDk,
- Guido Germano, PhDa,
- Yuka Otaki, MDa,
- Balaji K. Tamarappoo, MDa,
- Damini Dey, PhDa,
- Daniel S. Berman, MDa and
- Piotr J. Slomka, PhDa,∗ ()
- aDepartment of Imaging (Division of Nuclear Medicine), Medicine, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California
- bDepartment of Nuclear Cardiology, Assuta Medical Centers, Tel Aviv, Israel
- cBen Gurion University of the Negev, Beer Sheba, Israel
- dDivision of Cardiology, Department of Medicine, Columbia University Medical Center and New York-Presbyterian Hospital, New York, New York
- eDepartment of Radiology, Columbia University Medical Center and New York-Presbyterian Hospital, New York, New York
- fOregon Heart and Vascular Institute, Sacred Heart Medical Center, Springfield, Oregon
- gDivision of Cardiology, University of Ottawa Heart Institute, Ottawa, Canada
- hDepartment of Nuclear Medicine, Cardiac Imaging, University Hospital Zurich, Zurich, Switzerland
- iSection of Cardiovascular Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut
- jCardiovascular Imaging Technologies LLC, Kansas City, Missouri
- kDepartment of Radiology, Division of Nuclear Medicine and Molecular Imaging, Brigham and Women's Hospital, Boston, Massachusetts
- ↵∗Address for correspondence:
Dr. Piotr Slomka, Cedars-Sinai Medical Center, 8700 Beverly Boulevard, Suite A047N, Los Angeles, California 90048.
Objectives The study evaluated the automatic prediction of obstructive disease from myocardial perfusion imaging (MPI) by deep learning as compared with total perfusion deficit (TPD).
Background Deep convolutional neural networks trained with a large multicenter population may provide improved prediction of per-patient and per-vessel coronary artery disease from single-photon emission computed tomography MPI.
Methods A total of 1,638 patients (67% men) without known coronary artery disease, undergoing stress 99mTc-sestamibi or tetrofosmin MPI with new generation solid-state scanners in 9 different sites, with invasive coronary angiography performed within 6 months of MPI, were studied. Obstructive disease was defined as ≥70% narrowing of coronary arteries (≥50% for left main artery). Left ventricular myocardium was segmented using clinical nuclear cardiology software and verified by an expert reader. Stress TPD was computed using sex- and camera-specific normal limits. Deep learning was trained using raw and quantitative polar maps and evaluated for prediction of obstructive stenosis in a stratified 10-fold cross-validation procedure.
Results A total of 1,018 (62%) patients and 1,797 of 4,914 (37%) arteries had obstructive disease. Area under the receiver-operating characteristic curve for disease prediction by deep learning was higher than for TPD (per patient: 0.80 vs. 0.78; per vessel: 0.76 vs. 0.73: p < 0.01). With deep learning threshold set to the same specificity as TPD, per-patient sensitivity improved from 79.8% (TPD) to 82.3% (deep learning) (p < 0.05), and per-vessel sensitivity improved from 64.4% (TPD) to 69.8% (deep learning) (p < 0.01).
Conclusions Deep learning has the potential to improve automatic interpretation of MPI as compared with current clinical methods.
- convolutional neural network
- deep learning
- obstructive coronary artery disease
- SPECT myocardial perfusion imaging
This research was supported in part by grant R01HL089765 from the National Heart, Lung, and Blood Institute/National Institutes of Health (principal investigator: Dr. Slomka). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Dr. Sharir has served as a consultant for Spectrum Dynamics. Dr. Einstein has served as a consultant to GE Healthcare; and his institution has received research support from Toshiba America Medical Systems. Dr. Ruddy has received research grant support from GE Healthcare and Advanced Accelerator Applications. Dr. Miller has served as a consultant for GE, Bracco, and Alnylam; and has received grant support from Bracco. Dr. Dorbala owns stock in GE Healthcare. Dr. Di Carli has received research grant support from Spectrum Dynamics and consulting honoraria from Sanofi and GE Healthcare. Drs. Germano, Berman, and Slomka participate in software royalties for QPS software at Cedars-Sinai Medical Center. Dr. Slomka has received research grant support from Siemens Medical Systems. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Received November 2, 2017.
- Revision received January 15, 2018.
- Accepted January 16, 2018.
- 2018 American College of Cardiology Foundation