A decoupled dual-source learning framework trains parallel models on independent expert annotations for PET-CT bone infection segmentation and uses patient-level 3D evaluation to report performance variations.
Medl-u: Uncertainty-aware 3d automatic annotation based on evidential deep learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Cross-Source Supervision for Bone Infection Segmentation in Dual-Modality PET-CT
A decoupled dual-source learning framework trains parallel models on independent expert annotations for PET-CT bone infection segmentation and uses patient-level 3D evaluation to report performance variations.