Generalizable CT-free PET attenuation and scatter correction via few-shot cross domain adaptation

Authors:

Abstract
The rapid advancements in PET technology, coupled with the need for accurate and efficient imaging, necessitate the development of robust and generalizable methods for CT-free attenuation and scatter correction (ASC). Deep learning offers a promising solution, but exhibits limited performance when tested in diverse clinical settings and varying imaging conditions. We propose a few-shot fine-tuning paradigm that enables efficient adaptation of models from a source domain to a new target domain. Our backbone network incorporates statistical modulation to extract domain-specific distribution information and employs pixel-wise factor scaling modeling to disentangle ASC factor maps from input images. On a large and diverse dataset of 1539 subjects across multiple tracers, scanners, and centers, we evaluate model performance under single-tracer training, multi-tracer joint training, and few-shot adaptation strategies. Although joint training demonstrates strong performance on known tracers, the proposed few-shot adaptation approach, CrossPET-Adapt, excels at adapting to unseen domains with minimal data, outperforming joint training. This method significantly reduces radiation exposure and data requirements, offering a rapid and robust solution for CT-free PET ASC in varied clinical environments.
Funding Information
  • Shanghai Magnolia Talent Plan Pujiang Project (25PJD082)
  • National Natural Science Foundation of China (62471299)
  • National Natural Science Foundation of China (82372038)
  • National Key Research and Development Program of China
  • Natural Science Foundation of Guangdong Province in China (2023B1515120007)
  • Shen- zhen Science, the Technology Program of China (KJZD20240903101307010)
  • National Natu- ral Science Foundation of China project (UNNC Project ID B0166)
  • Yongjiang Technology Innovation Project (2022A-097-G)
  • Shanghai Municipal Key Clinical Specialty (shslczdzk03403)

This publication has 56 references indexed in Scilit: