Transfer Risk Map: Mitigating Pixel-level Negative Transfer in Medical Segmentation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912219382415360 |
|---|---|
| author | Duan, Shutong Yang, Jingyun Tan, Yang Zhang, Guoqing Li, Yang Zhang, Xiao-Ping |
| author_facet | Duan, Shutong Yang, Jingyun Tan, Yang Zhang, Guoqing Li, Yang Zhang, Xiao-Ping |
| contents | How to mitigate negative transfer in transfer learning is a long-standing and challenging issue, especially in the application of medical image segmentation. Existing methods for reducing negative transfer focus on classification or regression tasks, ignoring the non-uniform negative transfer risk in different image regions. In this work, we propose a simple yet effective weighted fine-tuning method that directs the model's attention towards regions with significant transfer risk for medical semantic segmentation. Specifically, we compute a transferability-guided transfer risk map to quantify the transfer hardness for each pixel and the potential risks of negative transfer. During the fine-tuning phase, we introduce a map-weighted loss function, normalized with image foreground size to counter class imbalance. Extensive experiments on brain segmentation datasets show our method significantly improves the target task performance, with gains of 4.37% on FeTS2021 and 1.81% on iSeg2019, avoiding negative transfer across modalities and tasks. Meanwhile, a 2.9% gain under a few-shot scenario validates the robustness of our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_02340 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Transfer Risk Map: Mitigating Pixel-level Negative Transfer in Medical Segmentation Duan, Shutong Yang, Jingyun Tan, Yang Zhang, Guoqing Li, Yang Zhang, Xiao-Ping Computer Vision and Pattern Recognition How to mitigate negative transfer in transfer learning is a long-standing and challenging issue, especially in the application of medical image segmentation. Existing methods for reducing negative transfer focus on classification or regression tasks, ignoring the non-uniform negative transfer risk in different image regions. In this work, we propose a simple yet effective weighted fine-tuning method that directs the model's attention towards regions with significant transfer risk for medical semantic segmentation. Specifically, we compute a transferability-guided transfer risk map to quantify the transfer hardness for each pixel and the potential risks of negative transfer. During the fine-tuning phase, we introduce a map-weighted loss function, normalized with image foreground size to counter class imbalance. Extensive experiments on brain segmentation datasets show our method significantly improves the target task performance, with gains of 4.37% on FeTS2021 and 1.81% on iSeg2019, avoiding negative transfer across modalities and tasks. Meanwhile, a 2.9% gain under a few-shot scenario validates the robustness of our approach. |
| title | Transfer Risk Map: Mitigating Pixel-level Negative Transfer in Medical Segmentation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.02340 |