| Autor: |
Guo X; Yale University, New Haven, CT 06511, USA., Shi L; IBM Research, San Jose, CA 95120, USA., Chen X; Yale University, New Haven, CT 06511, USA., Zhou B; Yale University, New Haven, CT 06511, USA., Liu Q; Yale University, New Haven, CT 06511, USA., Xie H; Yale University, New Haven, CT 06511, USA., Liu YH; Yale University, New Haven, CT 06511, USA., Palyo R; Yale New Haven Hospital, New Haven, CT 06511, USA., Miller EJ; Yale University, New Haven, CT 06511, USA., Sinusas AJ; Yale University, New Haven, CT 06511, USA., Spottiswoode B; Siemens Medical Solutions USA, Inc., Knoxville, TN 37932, USA., Liu C; Yale University, New Haven, CT 06511, USA., Dvornek NC; Yale University, New Haven, CT 06511, USA. |
| Abstrakt: |
The rapid tracer kinetics of rubidium-82 ( 82 Rb) 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 82 Rb 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. |
