paper-with-me

홈 › Papers

TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation

2026-04-17 · Tristan Kirscher, Alexandra Ertl, Klaus Maier-Hein, Xavier Coubez, Philippe Meyer, Sylvain Faisan arxiv

Pancreatic ductal adenocarcinoma (PDAC) segmentation on contrast-enhanced CT is inherently ambiguous: inter-rater disagreement among experts reflects genuine uncertainty rather than annotation noise. Standard deep learning approaches assume a single ground truth, producing probabilistic outputs that can be poorly calibrated and difficult to interpret under such ambiguity. We present TwinTrack, a framework that addresses this gap through post-hoc calibration of ensemble segmentation probabilities to the empirical mean human response (MHR) -the fraction of expert annotators labeling a voxel as tumor. Calibrated probabilities are thus directly interpretable as the expected proportion of annotators assigning the tumor label, explicitly modeling inter-rater disagreement. The proposed post-hoc calibration procedure is simple and requires only a small multi-rater calibration set. It consistently improves calibration metrics over standard approaches when evaluated on the MICCAI 2025 CURVAS-PDACVI multi-rater benchmark.

📄 PDF Abstract BibTeX arXiv:2604.15950

Code (0)

등록된 구현이 없습니다.

Tasks

Medical Image Segmentation

Similar Papers 제목 키워드 기반

Attention-Based Prototype Calibration for Multi-Rater Few-Shot Medical Image Segmentation

2026-06-15 · Truong Vu, Minh Khoi Ho, Yutong Xie arxiv

Few-shot medical image segmentation methods typically assume a single ground-truth annotation, overlooking systematic variability across expert raters commonly observed in clinical datasets. We propose an attention-based…

Medical Image SegmentationPersonalized Segmentation

Leveraging Multi-Rater Annotations to Calibrate Object Detectors in Microscopy Imaging

2026-01-30 · Francesco Campi, Lucrezia Tondo, Ekin Karabati, Johannes Betge 외 arxiv

Deep learning-based object detectors have achieved impressive performance in microscopy imaging, yet their confidence estimates often lack calibration, limiting their reliability for biomedical applications. In this work…

Diagnostic Uncertainty Calibration: Towards Reliable Machine Predictions in Medical Domain

2020-07-03 · Takahiro Mimori, Keiko Sasada, Hirotaka Matsui, Issei Sato

We propose an evaluation framework for class probability estimates (CPEs) in the presence of label uncertainty, which is commonly observed as diagnosis disagreement between experts in the medical domain. We also formaliz…

Diagnostic

Multi-Rater Calibrated Segmentation Models

2026-05-04 · Meritxell Riera-Marín, Javier García López, Júlia Rodríguez-Comas, Miguel A. González Ballester 외 arxiv

Objective: Accurate probability estimates are essential for the safe deployment of medical image segmentation models in clinical decision-making. However, modern deep segmentation networks are often poorly calibrated, a …

Medical Image Segmentation

PEBS: Per-rater Empirical-Bayes Shrinkage for RLHF Reward-Model Calibration

2026-06-25 · Arnav Raj arxiv

Reward models for Reinforcement Learning from Human Feedback (RLHF) pool preferences across thousands of annotators and fit one global affine calibrator, collapsing raters with systematically different rating-scale offse…

Reinforcement Learning