Foundation Model-guided Iteratively Prompting and Pseudo-Labeling for Partially Labeled Medical Image Segmentation
This addresses the challenge of degraded segmentation performance due to incomplete annotations in clinical settings, though it is incremental as it builds on existing pseudo-labeling and foundation model methods.
The authors tackled the problem of partially labeled medical image segmentation, where only a subset of organs are annotated, by proposing IPnP, an iterative prompting and pseudo-labeling framework that improves segmentation performance and approaches fully labeled reference results on the AMOS dataset.
Automated medical image segmentation has achieved remarkable progress with fully labeled data. However, site-specific clinical priorities and the high cost of manual annotation often yield scans with only a subset of organs labeled, leading to the partially labeled problem that degrades performance. To address this issue, we propose IPnP, an Iteratively Prompting and Pseudo-labeling framework, for partially labeled medical image segmentation. IPnP iteratively generates and refines pseudo-labels for unlabeled organs through collaboration between a trainable segmentation network (specialist) and a frozen foundation model (generalist), progressively recovering full-organ supervision. On the public dataset AMOS with the simulated partial-label setting, IPnP consistently improves segmentation performance over prior methods and approaches the performance of the fully labeled reference. We further evaluate on a private, partially labeled dataset of 210 head-and-neck cancer patients and demonstrate our effectiveness in real-world clinical settings.