Liu Xiaofeng, Huang Yongsong, Marin Thibault, Vafay Eslahi Samira, Tiss Amal, Chemli Yanis, Johnson Keith A, El Fakhri Georges, Ouyang Jinsong
Dept. of Radiology and Biomedical Imaging, Yale University, New Haven, CT, USA.
Dept. of Biomedical Informatics and Data Science, Yale University, New Haven, CT, USA.
Proc IEEE Int Symp Biomed Imaging. 2025 Apr;2025. doi: 10.1109/isbi60581.2025.10980695. Epub 2025 May 12.
The to-be-denoised positron emission tomography (PET) volumes are inherent with diverse count levels, which imposes challenges for a unified model to tackle varied cases. In this work, we resort to the recently flourished prompt learning to achieve generalizable PET denoising with different count levels. Specifically, we propose dual prompts to guide the PET denoising in a divide-and-conquer manner, i.e., an explicitly count-level prompt to provide the specific prior information and an implicitly general denoising prompt to encode the essential PET denoising knowledge. Then, a novel prompt fusion module is developed to unify the heterogeneous prompts, followed by a prompt-feature interaction module to inject prompts into the features. The prompts are able to dynamically guide the noise-conditioned denoising process. Therefore, we are able to efficiently train a unified denoising model for various count levels, and deploy it to different cases with personalized prompts. We evaluated on 1940 low-count PET 3D volumes with uniformly randomly selected 13-22% fractions of events from 97 F-MK6240 tau PET studies. It shows our dual prompting can largely improve the performance with informed count-level and outperform the count-conditional model.
待去噪的正电子发射断层扫描（PET）体数据具有不同的计数水平，这给统一模型处理各种情况带来了挑战。在这项工作中，我们借助最近蓬勃发展的提示学习来实现具有不同计数水平的通用PET去噪。具体而言，我们提出了双重提示，以分而治之的方式指导PET去噪，即一个明确的计数水平提示来提供特定的先验信息，以及一个隐式的通用去噪提示来编码基本的PET去噪知识。然后，开发了一个新颖的提示融合模块来统一异构提示，接着是一个提示-特征交互模块将提示注入到特征中。这些提示能够动态地指导噪声条件下的去噪过程。因此，我们能够有效地训练一个针对各种计数水平的统一去噪模型，并通过个性化提示将其部署到不同的情况中。我们对来自97项F-MK6240 tau PET研究的1940个低计数PET 3D体数据进行了评估，这些数据均匀随机地选取了13%-22%的事件分数。结果表明，我们的双重提示能够在已知计数水平的情况下大幅提高性能，并且优于计数条件模型。
