Deep Learning-Based Attenuation Correction Improves Diagnostic Accuracy of Cardiac SPECT.
Aakash D Shanbhag, Robert J H Miller, Konrad Pieszko, Mark Lemley, Paul Kavanagh, Attila Feher, Edward J Miller, Albert J Sinusas, Philipp A Kaufmann, Donghee Han, Cathleen Huang, Joanna X Liang, Daniel S Berman, Damini Dey, Piotr J Slomka
Author Information
Aakash D Shanbhag: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Robert J H Miller: Department of Cardiac Sciences, University of Calgary, Calgary, Alberta, Canada.
Konrad Pieszko: Department of Interventional Cardiology and Cardiac Surgery, University of Zielona Góra, Zielona Góra, Poland.
Mark Lemley: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Paul Kavanagh: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Attila Feher: Section of Cardiovascular Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut; and.
Edward J Miller: Section of Cardiovascular Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut; and.
Albert J Sinusas: Section of Cardiovascular Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut; and.
Philipp A Kaufmann: Cardiac Imaging, Department of Nuclear Medicine, University Hospital Zurich, Zurich, Switzerland.
Donghee Han: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Cathleen Huang: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Joanna X Liang: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Daniel S Berman: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Damini Dey: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California.
Piotr J Slomka: Departments of Medicine (Division of Artificial Intelligence in Medicine), Imaging, and Biomedical Sciences, Cedars-Sinai Medical Center, Los Angeles, California; piotr.slomka@cshs.org.
To improve diagnostic accuracy, myocardial perfusion imaging (MPI) SPECT studies can use CT-based attenuation correction (AC). However, CT-based AC is not available for most SPECT systems in clinical use, increases radiation exposure, and is impacted by misregistration. We developed and externally validated a deep-learning model to generate simulated AC images directly from non-AC (NC) SPECT, without the need for CT. SPECT myocardial perfusion imaging was performed using Tc-sestamibi or Tc-tetrofosmin on contemporary scanners with solid-state detectors. We developed a conditional generative adversarial neural network that applies a deep learning model (DeepAC) to generate simulated AC SPECT images. The model was trained with short-axis NC and AC images performed at 1 site ( = 4,886) and was tested on patients from 2 separate external sites ( = 604). We assessed the diagnostic accuracy of the stress total perfusion deficit (TPD) obtained from NC, AC, and DeepAC images for obstructive coronary artery disease (CAD) with area under the receiver-operating-characteristic curve. We also quantified the direct count change among AC, NC, and DeepAC images on a per-voxel basis. DeepAC could be obtained in less than 1 s from NC images; area under the receiver-operating-characteristic curve for obstructive CAD was higher for DeepAC TPD (0.79; 95% CI, 0.72-0.85) than for NC TPD (0.70; 95% CI, 0.63-0.78; < 0.001) and similar to AC TPD (0.81; 95% CI, 0.75-0.87; = 0.196). The normalcy rate in the low-likelihood-of-coronary-disease population was higher for DeepAC TPD (70.4%) and AC TPD (75.0%) than for NC TPD (54.6%, < 0.001 for both). The positive count change (increase in counts) was significantly higher for AC versus NC (median, 9.4; interquartile range, 6.0-14.2; < 0.001) than for AC versus DeepAC (median, 2.4; interquartile range, 1.3-4.2). In an independent external dataset, DeepAC provided improved diagnostic accuracy for obstructive CAD, as compared with NC images, and this accuracy was similar to that of actual AC. DeepAC simplifies the task of artifact identification for physicians, avoids misregistration artifacts, and can be performed rapidly without the need for CT hardware and additional acquisitions.
