Five modality-specific CLIP-like medical models are aligned through a shared, distilled text embedding space, enabling zero-shot cross-modal retrieval and improved few-shot classification without paired image data.
Quantifying tumour heterogeneity in 18 f-fdg pet/ct imaging by texture analysis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.IV 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Multimodal Medical Image Binding via Shared Text Embeddings
Five modality-specific CLIP-like medical models are aligned through a shared, distilled text embedding space, enabling zero-shot cross-modal retrieval and improved few-shot classification without paired image data.