A Multi-Cohort Validation of Censoring-Aware Conformal Lower Predictive Bounds for Pathology Survival Models

2026/07/23 by Mingi Hong
Mathematics · Computer Science · #stat.ME #cs.LG

paper · pdf

22 pages, 3 figures, 2 main tables, and supplementary material. Preprint

arxiv created 2026/07/23 · arxiv updated 2026/08/06

Abstract

Whole-slide survival models commonly provide risk rankings without calibrated statements about individual event times. We evaluate fixed-cutoff drcosarc, a post-hoc conformal wrapper for discrete-time multiple-instance learning survival heads using frozen UNI2-h representations, in an internal 18-configuration sweep across five TCGA cohorts and an external five-configuration evaluation across three CPTAC cohorts. We distinguish configuration--fold--split summaries of the inverse-probability-of-censoring-weighted (IPCW) estimate and median lower predictive bound (LPB) from a hierarchy-aware patient-ensemble estimand of the mean drcosarc--naive LPB difference. At α=0.1, the drcosarc IPCW estimate was nearest 0.90 in KIRC, LUAD, and STAD. Patient-ensemble drcosarc--naive intervals excluded zero in KIRC, KIRP, STAD, UCEC, and CPTAC-CCRCC, but included zero in internal LUAD, CPTAC-LUAD, CPTAC-UCEC, and the internal LUSC extension. In a 20-replicate low-censoring semi-synthetic setting with known event times, drcosarc empirical coverage was 0.9129 [0.9053, 0.9207]. An exploratory analysis supported a head-error-by-censoring interaction within that data-generating process. In a two-cohort ABMIL sensitivity analysis, increasing the hazard grid to K=16 raised localized marginal IPCW estimates above the prespecified 0.87 threshold and yielded positive paired LPB differences, although worst-group estimates remained below 0.87. Overall, performance was cohort dependent, and its interpretation changed with the patient-level unit, estimand, and censoring assumptions.

Citations

• Benchmarking Pathology Foundation Models for Breast Cancer Survival Prediction
• DPsurv: Dual-Prototype Evidential Fusion for Uncertainty-Aware and Interpretable Whole-Slide Image Survival Prediction
• PTCMIL: Multiple Instance Learning via Prompt Token Clustering for Whole Slide Image Analysis
• Conformal Survival Bands for Risk Screening under Right-Censoring
• Generating crossmodal gene expression from cancer histopathology improves multimodal AI predictions
• Doubly Robust Conformalized Survival Analysis with Right-Censored Data
• 2DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classification
• Mamba2MIL: State Space Duality Based Multiple Instance Learning for Computational Pathology
• Conformal predictive intervals in survival analysis: a re-sampling approach
• DGR-MIL: Exploring Diverse Global Representation in Multiple Instance Learning for Whole Slide Image Classification
• AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classification
• Conformalized Survival Distributions: A Generic Post-Process to Increase Calibration
• MambaMIL: Enhancing Long Sequence Modeling with Sequence Reordering in Computational Pathology
• Attention-Challenging Multiple Instance Learning for Whole Slide Image Classification
• Conformalized survival analysis with adaptive cutoffs
• DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification
• Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networks
• TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification
• Conformalized Survival Analysis
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning
• Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Images
• Continuous and Discrete-Time Survival Prediction with Neural Networks
• Effective Ways to Build and Evaluate Individual Survival Distributions