What question did this study set out to answer?

The aim is to develop a robust method for CT-free attenuation and scatter correction using few-shot adaptation.

May 16, 2026npj Digital MedicineOpen Access

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

View Full Paper

Ask AI

Bookmark

Authors

HWHanzhong WangShanghai Jiao Tong University MWMeiyuan WenUniversity of Nottingham Ningbo China XQXiaoya QiaoShanghai Jiao Tong University

Discussion

Member takes

Implication

Randomized trial demonstrates efficient PET attenuation and scatter correction in diverse imaging conditions, indicating promising advancements in clinical imaging methods.

Key Points

The aim is to develop a robust method for CT-free attenuation and scatter correction using few-shot adaptation.
Utilized a backbone network incorporating statistical modulation and pixel-wise factor scaling.
Implemented few-shot fine-tuning from a source domain to new target domains.
Evaluated performance on a dataset of 1539 subjects across various scanners and tracers.
Few-shot adaptation method, CrossPET-Adapt, outperformed joint training on unseen domains.
Significantly reduced radiation exposure during imaging.
Enhanced adaptation with minimal data requirements compared to traditional methods.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/6a080a41a487c87a6a40c2ee https://doi.org/10.1038/s41746-026-02760-w

View Full Paper

Ask AI

Bookmark

Implication

Randomized trial demonstrates efficient PET attenuation and scatter correction in diverse imaging conditions, indicating promising advancements in clinical imaging methods.

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

Discussion

Member takes

Implication

Key Points

Cite This Study

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

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider