# TwinTrack：面向医学图像分割的后验多标注者校准

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-04-17 08:00
- AIHOT 链接：https://aihot.virxact.com/items/cmo6z5esj00mpsli5jdv5aby2
- 原文链接：https://arxiv.org/abs/2604.15950

## AI 摘要

胰腺导管腺癌CT分割存在专家标注分歧，标准深度学习假设单一真值导致概率校准失真。TwinTrack框架通过将集成分割概率校准至经验平均人类响应(MHR)，使输出概率可直接解释为标注肿瘤的专家比例，显式量化标注不确定性。该方法仅需少量多标注者校准集，在MICCAI 2025 CURVAS-PDACVI基准上持续改进校准指标，为医学影像分割提供可解释的不确定性建模方案。

## 正文

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.