TiDE-Net: A time-guided dual-encoder ResUNet for Positron Emission Tomography (PET) image denoising
Our experiments demonstrate that conditioning on a timestep enables TiDE-Net to generalize across unseen noise levels, achieving consistent denoising performance across different noise conditions.
Authors
- Lei Bi
- Yuxin Xue (ORCID: https://orcid.org/0009-0005-1361-8162)
- Michael Fulham (ORCID: https://orcid.org/0000-0003-0602-6319)
- Jinman Kim (ORCID: https://orcid.org/0000-0001-5960-1060)
- Bo Tan (ORCID: https://orcid.org/0009-0005-3570-5857)
Institutions
- The University of Sydney (AU)
- Royal Prince Alfred Hospital (AU)
Publication Details
- Journal
- Computer Methods and Programs in Biomedicine
- Published
- 2025-12-03
- DOI
- https://doi.org/10.1016/j.cmpb.2025.109200
- Primary Topic
- Medical Imaging Techniques and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00