Myocardial perfusion imaging (MPI) using single photon emission computed tomography (SPECT) is a prevalent clinical tool in diagnosis and prognosis of coronal artery disease (CAD). Dual cardiac-respiratory gated SPECT (D-SPECT) has been proposed long ago to eliminate artifacts from patient's cardiac and respiratory motion, but elevated noise and lack of evidence for diagnostic benefit over processing complexity hamper its clinical adoption. Cardiac ECG-gated SPECT (C-SPECT) remains the mainstream enabling the assessment of left-ventricular (LV) functions. Advanced 4D reconstruction with superior denoising ability resurrects interests in D-SPECT recently. Meanwhile, attenuation correction (AC) is essential for quantitative MPI and minimizing false positives. However, AC artifacts can arise when SPECT images are mismatched with attenuation maps derived from the hybrid SPECT/CT scanner. This becomes another major hurdle for D- SPECT because the accompanied cone-beam CT (CBCT) yields only one non-respiratory-gated 3D CT image. To address these issues, we propose a novel SPECT MPI protocol: D-SPECT with motion matched AC (D- SPECT-MAC), which will minimize SPECT MPI artifacts. The motion-matched attenuation maps can be obtained by a general simultaneous motion estimation and image reconstruction (G-SMEIR) method and D- SPECT images without changing low-dose CBCT imaging. Founded on our success on 4D spatiotemporal reconstruction and artifact-correction methods for both SPECT and CBCT and experience collecting D-SPECT data for more than 1,500 patients, with advances in development of realistic computational human phantoms and computer power, it is time to systematically evaluate D-SPECT-MAC using a mass simulation study backed by a patient study to provide the urgently needed evidence for use of D-SPECT. We hypothesize that D-SPECT-MAC can significantly improve perfusion defect detection and LV functional assessment, compared to two other protocols: popular C-SPECT with averaged AC over respiratory cycles (C-SPECT-AAC) and emerging D-SPECT with averaged AC (D-SPECT-AAC). Our hypothesis will be tested through the following specific aims: 1) To evaluate performance of three SPECT MPI protocols using a mass simulation study of 5,000 virtual patients from the XCAT phantom program; 2) To develop D-SPECT-MAC for clinical data and to verify simulation findings using 150 patient data sets retrospectively selected. The success of this project will lead to artifact-free SPECT MPI, which enables use of only one set of faithfully corrected MPI images instead of visual reading both non-corrected and AC images in the current clinical practice, thus significantly improving diagnostic accuracy, reproducibility, and clinical workflow. Furthermore, this project will provide opportunities for under-represented students at the PI's institute to be professionally trained and to gain precious experience on medical imaging research with clinical collaborators at medical institutes, which paves a way for them to pursue a successful biomedical career.

Public Health Relevance

The proposed research is highly relevant to public health because the development of respiratory motion matched attenuation correction using low-dose X-ray cone-beam computed tomography (CBCT) for myocardial perfusion imaging using single photon emission computed tomography (SPECT) could substantially reduce image artifacts and improve diagnostic accuracy for heart disease. SPECT is a slow imaging modality that is prone to motion and attenuation artifacts. The proposed method that holds the potential to eliminate these artifacts using advanced image reconstruction and processing techniques will be evaluated cost-effectively by mass computer simulations and a preliminary patient study.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Academic Research Enhancement Awards (AREA) (R15)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1)
Program Officer
Danthi, Narasimhan
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Texas Arlington
Schools of Nursing
United States
Zip Code