TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918508785303552 |
|---|---|
| author | Kirscher, Tristan Ertl, Alexandra Maier-Hein, Klaus Coubez, Xavier Meyer, Philippe Faisan, Sylvain |
| author_facet | Kirscher, Tristan Ertl, Alexandra Maier-Hein, Klaus Coubez, Xavier Meyer, Philippe Faisan, Sylvain |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_15950 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation Kirscher, Tristan Ertl, Alexandra Maier-Hein, Klaus Coubez, Xavier Meyer, Philippe Faisan, Sylvain Machine Learning 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. |
| title | TwinTrack: Post-hoc Multi-Rater Calibration for Medical Image Segmentation |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2604.15950 |