Proceedings

Bridging the Objective Gap: A Unified Pre-Training Framework for Few-Shot Medical Image Segmentation

Shoupeng Chen, Yiming Miao, Limei Peng, Pin-Han Ho

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 998-1006.

https://doi.org/10.24963/ijcai.2026/112

Keywords

  • Machine Learning: Few-shot learning
  • Computer Vision: Biomedical image analysis
  • Computer Vision: Segmentation, grouping and shape analysis

Abstract

Few-shot medical image segmentation relies on dense, boundary-sensitive prototype matching, yet common pre-training objectives mainly optimize global alignment or reconstruction, creating an objective gap that hurts boundary delineation and increases adaptation cost. This raises the question: how to pre-train representations intrinsically matchable for episodic FSS while requiring minimal adaptation-induced re-organization? We introduce a Drift-Gap diagnostic to quantify intrinsic dense-matching misalignment and adaptation-induced feature drift. Guided by this lens, we propose BOG-PRETRAIN, combining Reliability-Gated Alignment to mitigate noisy report supervision, Semantic-Guided MIM to emphasize boundary-informative regions, and Dual Consistency Regularization to stabilize episodic metric geometry. Across five benchmarks (1/4/16-shot), BOG-PRETRAIN improves mean Dice by +15.5/+15.9/+12.5 points over best priors and reduces mean HD95 by 0.9/4.6/10.1; it achieves the lowest Drift (0.052 vs. 0.155 baseline) and Gap_PT (0.256), with ablations confirming the components' complementarity.

Shoupeng Chen, Yiming Miao, Limei Peng, and Pin-Han Ho. 2026. Bridging the Objective Gap: A Unified Pre-Training Framework for Few-Shot Medical Image Segmentation. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 998–1006. https://doi.org/10.24963/ijcai.2026/112
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards