mlsanityData to 7 Oct 2026

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

arXiv 2608.04025 ↗First posted 23 Jul 2026stat.ME cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 26 of its 39 references that are in the map. First posted on arXiv 23 Jul 2026.

Builds on

26 of 39 references.

References · 26

  1. 2020Data Efficient and Weakly Supervised Computational Pathology on Whole Slide Imagescited by 738
  2. 2021TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classificationcited by 549
  3. 2020Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learningcited by 473
  4. 2022DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classificationcited by 254
  5. 2021Whole Slide Images are 2D Point Clouds: Context-Aware Survival Prediction using Patch-based Graph Convolutional Networkscited by 124
  6. 2024MambaMIL: Enhancing Long Sequence Modeling with Sequence Reordering in Computational Pathologycited by 87
  7. 2019Continuous and Discrete-Time Survival Prediction with Neural Networkscited by 68
  8. 2018Effective Ways to Build and Evaluate Individual Survival Distributionscited by 64
  9. 2023Attention-Challenging Multiple Instance Learning for Whole Slide Image Classificationcited by 61
  10. 2021Conformalized Survival Analysiscited by 53
  11. 20242DMamba: Efficient State Space Model for Image Representation with Applications on Giga-Pixel Whole Slide Image Classificationcited by 35
  12. 2024DGR-MIL: Exploring Diverse Global Representation in Multiple Instance Learning for Whole Slide Image Classificationcited by 25
14 more
  1. 2022Conformalized survival analysis with adaptive cutoffscited by 17
  2. 2024Conformalized Survival Distributions: A Generic Post-Process to Increase Calibrationcited by 10
  3. 2024AEM: Attention Entropy Maximization for Multiple Instance Learning based Whole Slide Image Classificationcited by 10
  4. 2025DPsurv: Dual-Prototype Evidential Fusion for Uncertainty-Aware and Interpretable Whole-Slide Image Survival Predictioncited by 6
  5. 2024Doubly Robust Conformalized Survival Analysis with Right-Censored Datacited by 5
  6. 2024Mamba2MIL: State Space Duality Based Multiple Instance Learning for Computational Pathologycited by 4
  7. 2025Generating crossmodal gene expression from cancer histopathology improves multimodal AI predictionscited by 4
  8. 2025Conformal Survival Bands for Risk Screening under Right-Censoringcited by 4
  9. 2024Conformal predictive intervals in survival analysis: a re-sampling approachcited by 2
  10. 2025PTCMIL: Multiple Instance Learning via Prompt Token Clustering for Whole Slide Image Analysiscited by 2
  11. 2026Good Rankings, Wrong Probabilities: A Calibration Audit of Multimodal Cancer Survival Modelscited by 1
  12. 2026MambaBack: Bridging Local Features and Global Contexts in Whole Slide Image Analysiscited by 1
  13. 2026Benchmarking Pathology Foundation Models for Breast Cancer Survival Predictioncited by 1
  14. 2026History-Aware Conformal Prediction Sets for Censored Time-to-Event Outcomescited by 1

Built on by

Not cited yet.