CMGL separates modality confidence estimation via evidential deep learning from graph-based fusion to improve cancer subtype classification accuracy by 4.03% on average and recover known subtypes like PAM50 in BRCA with transfer to KIRC.
Comprehensive molecular characterization of clear cell renal cell carcinoma
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
CMGL: Confidence-guided Multi-omics Graph Learning for Cancer Subtype Classification
CMGL separates modality confidence estimation via evidential deep learning from graph-based fusion to improve cancer subtype classification accuracy by 4.03% on average and recover known subtypes like PAM50 in BRCA with transfer to KIRC.