CIGTSurv uses clinical text embeddings to guide cross-attention and distribution alignment between pathology and genomics, reaching an average C-index of 0.788 across five TCGA cohorts.
Pan-cancer integrative histology-genomic analysis via multimodal deep learning
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
CIGTSurv: Clinical Information Guided Tri-modal Survival Prediction with Local Prototype Association and Global Feature Alignment
CIGTSurv uses clinical text embeddings to guide cross-attention and distribution alignment between pathology and genomics, reaching an average C-index of 0.788 across five TCGA cohorts.