# A foundation model-driven multi-view collaborative framework for semi-supervised 3D medical image segmentation

## TL;DR

This paper introduces a new framework for 3D medical image segmentation that uses foundation models and multi-view learning to reduce the need for labeled data.

## Contribution

The novel contribution is a multi-view collaborative framework leveraging SAM-like foundation models for semi-supervised 3D segmentation.

## Key findings

- The proposed method outperforms existing SAM-based semi-supervised approaches on MRI brain tumor and PET heart segmentation datasets.
- Multi-view collaboration improves boundary precision and shows strong modality transferability.
- The framework reduces annotation costs while maintaining high segmentation accuracy.

## Abstract

3D medical image segmentation is a cornerstone for quantitative analysis and clinical decision-making in various modalities. However, acquiring high-quality voxel-level annotations is both time-consuming and labor-intensive. Semi-supervised learning (SSL) provides an appealing solution by effectively utilizing limited labeled data along with abundant unlabeled data to enhance segmentation performance under clinical data constraints.

We propose a foundation model-driven multi-view collaborative learning framework that exploits zero-shot capabilities of SAM-like foundation models to jointly learn from axial, sagittal, and coronal planes. A collaborative fusion module integrates complementary representations across views, enhancing 3D structural understanding and improving the performance with limited annotation cost.

Extensive experiments on two evaluation datasets including MRI brain tumor segmentation and whole-body PET heart segmentation demonstrate that our proposed method consistently outperforms existing SAM-based semi-supervised approaches. The multi-view collaborative design not only refines boundary precision for organ and tumor delineation but also shows strong transferability across imaging modalities.

This study presents a foundation model-driven, multi-view collaborative learning paradigm that efficiently advances semi-supervised 3D medical image segmentation, which provides a scalable and clinically meaningful solution that reduces annotation dependency while maintaining high segmentation accuracy across diverse medical imaging modalities.