Radiation Dose Reduction in CT Myocardial Perfusion Imaging Using SMART-RECON

Area of Science:

• Medical Imaging
• Radiology
• Image Reconstruction

Background:

• Low-dose CT myocardial perfusion imaging (CT-MPI) is crucial for diagnosing heart conditions.
• Conventional CT-MPI often requires higher radiation doses to achieve adequate image quality.
• Existing reconstruction methods struggle with noise and artifact reduction at low doses.

Purpose of the Study:

• To introduce and validate a novel statistical model-based image reconstruction method, SMART-RECON, for low-dose CT-MPI.
• To evaluate the performance of SMART-RECON in terms of noise reduction, quantitative accuracy, and image quality compared to existing methods.
• To assess the dose and contrast independence of SMART-RECON's image properties.

Main Methods:

• Application of SMART-RECON, utilizing the nuclear norm of the spatial-temporal image matrix as a regularizer.
• Validation through numerical simulations and in vivo animal studies (porcine heart).
• Comparison with conventional filtered backprojection and spatial-temporal total variation (ST-TV-SIR) methods.

Main Results:

• SMART-RECON maintained quantitative accuracy of perfusion maps at radiation doses as low as 10 mAs, a significant reduction from the 200 mAs reference.
• Flow-occluded myocardium was well delineated at 10 mAs, outperforming TV-SIR and ST-TV-SIR.
• Image sharpness in SMART-RECON was largely independent of contrast and dose levels, unlike TV-SIR and ST-TV-SIR.
• Dose-normalized noise power spectrum (NPS) structure was dose-independent for SMART-RECON, but dose-dependent for TV-SIR and ST-TV-SIR.

Conclusions:

• SMART-RECON effectively reduces noise and artifacts in low-dose CT-MPI.
• The method preserves quantitative accuracy and image sharpness across varying dose and contrast levels.
• SMART-RECON offers a promising advancement for safer and more effective CT-MPI diagnostics.