Bio2Vol

---

Official codebase for MICCAI 2025 paper: Bio2Vol

News

• [25/09] Initialize the codebase.
• [25/09] This paper is accepted by MICCAI 2025.

Abstract

2D biomedical foundation models (FM) have demonstrated remarkable capabilities in 2D medical image segmentation across various modalities, with text-prompted approaches offering scalable analysis that facilitate integration with LLMs and clinical application. Adapting these models for 3D medical image segmentation can leverage their rich visual features while enabling text-prompted volumetric image segmentation. However, efficient adaptation poses significant challenges due to the substantial disparity between 2D and 3D medical images and the necessity to establish text-volume alignment. To address these limitations, we propose Bio2Vol, a novel adaptation framework that enables text-prompted 2D biomedical FMs to effectively handle volumetric data. Specifically, (1) To bridge the dimensional disparity, we propose a DualRate Sampling strategy (DRS) that processes inter slices within a volume at both sparse and dense intervals, capturing global contexts and local details; (2) To enhance volumetric feature representation, a Crossslice Dual-head Attention (CSDHA) is built upon the intra-slice features by repurposing existing pre-trained attention modules for parameterefficient inter-slice information fusion; and (3) To establish text-volume understanding, a Semantic Text-Visual Alignment loss (SAT) is used to extend the existing 2D text-visual alignment to the volumetric domain. Using BiomedParse as a demonstration case, extensive evaluation across 11 medical datasets across diverse anatomical regions and modalities shows that Bio2Vol significantly improves 3D medical image segmentation performance, enhancing DSC by 4.72% on Amos22 dataset with substantial improvements across MSD tasks.

Usage

---

0. Requirements

• Python 3.8+
• PyTorch 2.0.1
• MONAI 1.0.0
• CUDA 11.8
• cuDNN 8.5
• NVIDIA GPU with compute capability 8.6

1. Datasets

to do...

2. Training

to do...

3. Evaluation

to do...

4. Model Usage

to do...

Reference

If you find this repo useful for your research, please consider citing the paper as follows:

@inproceedings{bio2vol,
  title={Bio2Vol: Adapting 2D Biomedical Foundation Models for Volumetric Medical Image Segmentation},
  author={Zhuang, Jiaxin and Wu, Linshan and Ni, Xuefeng and Wang, Xi and Wang, Liansheng and Chen, Hao},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  year={2025},
  organization={Springer}
}