ProSeg models multi-rater lesion segmentation with two latent variables for expert preferences and boundary ambiguity using variational inference, achieving state-of-the-art results on NPC and LIDC-IDRI datasets while enabling both diversity and personalization.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Probabilistic Modeling of Multi-rater Medical Image Segmentation for Diversity and Personalization
ProSeg models multi-rater lesion segmentation with two latent variables for expert preferences and boundary ambiguity using variational inference, achieving state-of-the-art results on NPC and LIDC-IDRI datasets while enabling both diversity and personalization.