There is a newer version of the record available.

Published October 7, 2026 | Version v4.0

cpath-triage: diagnostic analysis of verbalized confidence in a medical vision-language model

  • 1. Department of Computer and Information Science and Engineering, University of Florida, Gainesville, FL, U.S.A.
  • 2. Department of Computer Science and Engineering, JK Lakshmipat University, Jaipur, India
  • 3. Department of Data Science and Engineering, Manipal University Jaipur, Jaipur, India
  • 4. Department of Computer Applications, Manipal University Jaipur, Jaipur, India

Description

Evaluation and analysis code, plus the prediction artifacts behind every table and figure, for a diagnostic study of verbalized confidence in MedGemma-27b-it on colorectal H&E histopathology patches. Covers zero-shot classification, temperature scaling and calibration metrics, selective-prediction routing with explicit tie handling, a design-effect sensitivity analysis in place of clustered intervals, and K=5 prompt-variant consistency routing.

Version 4.0 adds a cohort-weighted re-analysis of the K=5 subsets, a wording-only prompt family, image-removal, image-swap and pixel-shuffle controls, and a PLIP image-text baseline.

Notes

If you use this software or the prediction artifacts it contains, please cite both the software and the paper below.

Files

cpath-triage-v4.0-code-and-artifacts.zip

Files (22.2 MB)

Name Size
md5:a4db61b78385b20949c47be9bef7d933
22.2 MB Preview Download

Additional details

Related works