New survival model validation method shows cohort-dependent performance

Researchers have developed and validated a method called drcosarc, a post-hoc conformal wrapper for discrete-time multiple-instance learning survival models. This method was tested across multiple cohorts from The Cancer Genome Atlas (TCGA) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC). The evaluation focused on the inverse-probability-of-censoring-weighted (IPCW) estimate and median lower predictive bound (LPB), with results showing performance varied depending on the cohort, patient-level unit, estimand, and censoring assumptions. AI

IMPACT This research introduces a new validation technique for survival models, potentially improving their reliability in clinical settings.

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

Read on arXiv cs.LG →

• ABMIL
• CPTAC-CCRCC
• CPTAC-LUAD
• CPTAC-UCEC
• drcosarc
• Kirpal Nandra
• Stein Amortized Divergence
• The Cancer Genome Atlas
• Ucechukwu Iheadindu
• UNI2-h

• paper
• other

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