UNICON attaches task-specific adapters (LoRA, MLP, decoder, fusion) to a frozen CT foundation model, enabling continual extension to prognosis, segmentation, and PET scans.
In: 3D head and neck tumor segmentation in PET/CT challenge, pp
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.IV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
UNICON: UNIfied CONtinual Learning for Medical Foundational Models
UNICON attaches task-specific adapters (LoRA, MLP, decoder, fusion) to a frozen CT foundation model, enabling continual extension to prognosis, segmentation, and PET scans.