MRI-Guided Motion-Corrected PET Image Reconstruction for Cardiac PET/MRI

Simultaneous PET/MRI has shown potential for the comprehensive assessment of myocardial health from a single examination. Furthermore, MRI-derived respiratory motion information, when incorporated into the PET image reconstruction, has been shown to improve PET image quality. Separately, MRI-based anatomically guided PET image reconstruction has been shown to effectively denoise images, but this denoising has so far been demonstrated mainly in brain imaging. To date, the combined benefits of motion compensation and anatomic guidance have not been demonstrated for myocardial PET/MRI. This work addressed this lack by proposing a single cardiac PET/MR image reconstruction framework that fully utilizes MRI-derived information to allow both motion compensation and anatomic guidance within the reconstruction. Methods: Fifteen patients underwent an ¹⁸F-FDG cardiac PET/MRI scan with a previously introduced acquisition framework. The MRI data processing and image reconstruction pipeline produces respiratory motion fields and a high-resolution respiratory motion-corrected MR image with good tissue contrast. This MRI-derived information was then included in a respiratory motion-corrected, cardiac-gated, anatomically guided image reconstruction of the simultaneously acquired PET data. Reconstructions were evaluated by measuring myocardial contrast and noise and were compared with images from several comparative intermediate methods using the components of the proposed framework separately. Results: Including respiratory motion correction, cardiac gating, and anatomic guidance significantly increased contrast. In particular, myocardium-to-blood pool contrast increased by 143% on average (P < 0.0001), compared with conventional uncorrected, nonguided PET images. Furthermore, anatomic guidance significantly reduced image noise, by 16.1%, compared with nonguided image reconstruction (P < 0.0001). Conclusion: The proposed framework for MRI-derived motion compensation and anatomic guidance of cardiac PET data significantly improved image quality compared with alternative reconstruction methods. Each component of the reconstruction pipeline had a positive impact on the final image quality. These improvements have the potential to improve clinical interpretability and diagnosis based on cardiac PET/MR images.