Scribble-Supervised Medical Image Segmentation with Dynamic Teacher Switching and Hierarchical Consistency
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912844199493632 |
|---|---|
| author | Nguyen, Thanh-Huy Cao, Hoang-Loc Chung, Dat T. Vu, Mai-Anh Nguyen, Thanh-Minh Le, Minh Huynh, Phat K. Bagci, Ulas |
| author_facet | Nguyen, Thanh-Huy Cao, Hoang-Loc Chung, Dat T. Vu, Mai-Anh Nguyen, Thanh-Minh Le, Minh Huynh, Phat K. Bagci, Ulas |
| contents | Scribble-supervised methods have emerged to mitigate the prohibitive annotation burden in medical image segmentation. However, the inherent sparsity of these annotations introduces significant ambiguity, which results in noisy pseudo-label propagation and hinders the learning of robust anatomical boundaries. To address this challenge, we propose SDT-Net, a novel dual-teacher, single-student framework designed to maximize supervision quality from these weak signals. Our method features a Dynamic Teacher Switching (DTS) module to adaptively select the most reliable teacher. This selected teacher then guides the student via two synergistic mechanisms: high-confidence pseudo-labels, refined by a Pick Reliable Pixels (PRP) mechanism, and multi-level feature alignment, enforced by a Hierarchical Consistency (HiCo) module. Extensive experiments on the ACDC and MSCMRseg datasets demonstrate that SDT-Net achieves state-of-the-art performance, producing more accurate and anatomically plausible segmentation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_14563 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Scribble-Supervised Medical Image Segmentation with Dynamic Teacher Switching and Hierarchical Consistency Nguyen, Thanh-Huy Cao, Hoang-Loc Chung, Dat T. Vu, Mai-Anh Nguyen, Thanh-Minh Le, Minh Huynh, Phat K. Bagci, Ulas Computer Vision and Pattern Recognition Scribble-supervised methods have emerged to mitigate the prohibitive annotation burden in medical image segmentation. However, the inherent sparsity of these annotations introduces significant ambiguity, which results in noisy pseudo-label propagation and hinders the learning of robust anatomical boundaries. To address this challenge, we propose SDT-Net, a novel dual-teacher, single-student framework designed to maximize supervision quality from these weak signals. Our method features a Dynamic Teacher Switching (DTS) module to adaptively select the most reliable teacher. This selected teacher then guides the student via two synergistic mechanisms: high-confidence pseudo-labels, refined by a Pick Reliable Pixels (PRP) mechanism, and multi-level feature alignment, enforced by a Hierarchical Consistency (HiCo) module. Extensive experiments on the ACDC and MSCMRseg datasets demonstrate that SDT-Net achieves state-of-the-art performance, producing more accurate and anatomically plausible segmentation. |
| title | Scribble-Supervised Medical Image Segmentation with Dynamic Teacher Switching and Hierarchical Consistency |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.14563 |