B-MIM: Biased Masked Image Modeling for Generalizable Segmentation of Fine-Grained Anatomical Structures

arXiv:2608.24364v1 Announce Type: new Abstract: Self-supervised pretraining enables transferable representations for medical imaging, yet most CT encoders remain biased toward coarse semantic understanding, limiting their sensitivity to fine-grained anatomical structures such as vessels or small tumors. In this paper, we introduce Biased Masked Image Modeling (B-MIM), a modification of the iBOT objective that stochastically reduces global semantic alignment to prioritize local patch reconstruction. This bias encourages the encoder to capture high-frequency morphological details and structural continuity. We curate a multi-institutional CT abdominal dataset of 9,955 filtered studies from 17 public sources and pretrain a 3D Swin Transformer backbone using B-MIM. Across inter-dataset experiments on liver vessel segmentation, the proposed encoder improves topological fidelity (clDice) and achieves competitive Dice scores in tumor segmentation, compared to fully fine-tuned baselines, despite updating only a fraction of the parameters. Our results suggest that reducing global semantic pressure during pretraining enhances generalization to intricate anatomical structures.

Related

• Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imaging
• CA-GCL: Cross-Anatomy Global-Local Contrastive Learning for Robust 3D Medical Image Understanding
• Learning Coarse-to-Fine Osteoarthritis Representations under Noisy Hierarchical Labels
• Structure-Guided Self-Supervised Matching for One-Shot Medical Landmark Detection

Source: arXiv cs.CV | 2026-08-26