← Search

Zhanli Hu

2 accepted papers

2026

FlowPET: Physics-Informed Symplectic Flow Matching for Low-Count PET Reconstruction

ICML 2026poster

Low-count Positron Emission Tomography (PET) reconstruction is severely hindered by the dissipative nature of prevailing generative models, where the inherent phase-space contraction leads to the numerical extinction (``wash-out'') of weak but diagnostically critical lesion signals. To overcome this…

Cited by 0SourceScholar
2026

FourierPET: Deep Fourier-based Unrolled Network for Low-count PET Reconstruction

AAAI 2026technical

Low-count positron emission tomography (PET) reconstruction is a challenging inverse problem due to severe degradations arising from Poisson noise, photon scarcity, and attenuation correction errors. Existing deep learning methods typically address these in the spatial domain with an undifferentiate

Cited by 0SourcePDFScholar