New SHIFT model predicts survival from incomplete genomic data

Researchers have developed a novel missingness-aware survival model called SHIFT, designed to predict patient survival from incomplete and heterogeneous genomic data. Unlike existing methods that exclude or impute missing data, SHIFT directly utilizes observed inputs through masked self-attention and a feature-availability mask. The model was evaluated on glioblastoma and lung squamous cell carcinoma, demonstrating strong generalization and favorable comparisons against standard survival baselines, even with significant cross-cohort panel mismatches. AI

IMPACT This model could improve the accuracy and applicability of genomic survival predictions in multi-center precision oncology studies.

RANK_REASON The cluster contains a research paper detailing a new model and methodology. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• glioblastoma
• Gotit.pub
• Hugging Face
• Muhammet Sami Yavuz
• ScienceCast
• SHIFT
• squamous cell carcinoma of the lung
• Transformer++

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →