TAI-GAN: A Temporally and Anatomically Informed Generative Adversarial Network for early-to-late frame conversion in dynamic cardiac PET inter-frame motion correction.

Inter-frame motion in dynamic cardiac positron emission tomography (PET) using rubidium-82 (82Rb) myocardial perfusion imaging impacts myocardial blood flow (MBF) quantification and the diagnosis accuracy of coronary artery diseases. However, the high cross-frame distribution variation due to rapid tracer kinetics poses a considerable challenge for inter-frame motion correction, especially for early frames where intensity-based image registration techniques often fail. To address this issue, we propose a novel method called Temporally and Anatomically Informed Generative Adversarial Network (TAI-GAN) that utilizes an all-to-one mapping to convert early frames into those with tracer distribution similar to the last reference frame. The TAI-GAN consists of a feature-wise linear modulation layer that encodes channel-wise parameters generated from temporal information and rough cardiac segmentation masks with local shifts that serve as anatomical information. Our proposed method was evaluated on a clinical 82Rb PET dataset, and the results show that our TAI-GAN can produce converted early frames with high image quality, comparable to the real reference frames. After TAI-GAN conversion, the motion estimation accuracy and subsequent myocardial blood flow (MBF) quantification with both conventional and deep learning-based motion correction methods were improved compared to using the original frames. The code is available at https://github.com/gxq1998/TAI-GAN.

TAI-GAN: Temporally and Anatomically Informed GAN for Early-to-Late Frame Conversion in Dynamic Cardiac PET Motion Correction.

The rapid tracer kinetics of rubidium-82 (82Rb) and high variation of cross-frame distribution in dynamic cardiac positron emission tomography (PET) raise significant challenges for inter-frame motion correction, particularly for the early frames where conventional intensity-based image registration techniques are not applicable. Alternatively, a promising approach utilizes generative methods to handle the tracer distribution changes to assist existing registration methods. To improve frame-wise registration and parametric quantification, we propose a Temporally and Anatomically Informed Generative Adversarial Network (TAI-GAN) to transform the early frames into the late reference frame using an all-to-one mapping. Specifically, a feature-wise linear modulation layer encodes channel-wise parameters generated from temporal tracer kinetics information, and rough cardiac segmentations with local shifts serve as the anatomical information. We validated our proposed method on a clinical 82Rb PET dataset and found that our TAI-GAN can produce converted early frames with high image quality, comparable to the real reference frames. After TAI-GAN conversion, motion estimation accuracy and clinical myocardial blood flow (MBF) quantification were improved compared to using the original frames. Our code is published at https://github.com/gxq1998/TAI-GAN.

High-Sensitivity and High-Resolution SPECT/CT Systems Provide Substantial Dose Reduction Without Compromising Quantitative Precision for Assessment of Myocardial Perfusion and Function.

UNLABELLED: There is increasing concern about radiation exposure from myocardial perfusion SPECT (MPS). γ-cameras with solid-state cadmium-zinc-telluride (CZT) detectors have better count sensitivity and spatial resolution than conventional sodium iodine detectors, allowing for significant reductions in radiotracer dose or acquisition time. This study aimed to demonstrate the capability of a hybrid CZT SPECT/64-slice CT system for dose reduction and to determine the maximal reduction possible without compromising image quality or the quantification precision of clinical MPS. METHODS: The imaging data of patients with normal myocardial perfusion and 30 patients with mid-sized to large perfusion defects who had undergone stress (99m)Tc-tetrofosmin MPS were postprocessed. Low-dose (361 ± 60 MBq) and high-dose (725 ± 142 MBq) (99m)Tc-tetrofosmin scans were included, with 6-min and 4-min scanning times, respectively. List-mode SPECT data were reconstructed with CT-based attenuation correction and with full as well as 50% and 75% reductions in acquisition time to simulate the corresponding relative dose reductions. The reconstructed SPECT images were analyzed to calculate global MPS defect size and regional defect size for 3 coronary artery territories-left anterior descending, left circumflex, and right-as well as left ventricular (LV) volume and ejection fraction. RESULTS: For patients with normal MPS results, there were no differences in defect size, LV volume, or ejection fraction, regardless of whether 50% or 75% reduction was used. For patients with abnormal MPS results, at a 50% reduction there was a significant difference in global defect size but not in regional defect size in the left anterior descending, left circumflex, and right coronary artery territories, whereas at a 75% reduction the difference was statistically significant in all territories, including the difference in global defect size. Nonetheless, differences in the defect size were minimal. The LV end-diastolic and end-systolic volumes and LV ejection fraction were not significantly different, regardless of whether 50% or 75% dose reduction was used. CONCLUSION: Ultra-low-dose (<190 MBq) MPS even with short imaging times (<6 min) is feasible using a hybrid CZT SPECT/CT camera without compromising image quality or significantly altering quantification of myocardial perfusion or LV function. We demonstrated that an additional 50% reduction in the current low-dose recommendations from the American Society of Nuclear Cardiology guidelines for (99m)Tc-labeled MPS is highly feasible while retaining short imaging protocols.