UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation

Computer Science > Computer Vision and Pattern Recognition arXiv:2610.06938 (cs) [Submitted on 3 Oct 2026] Title: UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation.

计算机科学 > 计算机视觉与模式识别 arXiv:2610.06938 (cs) [提交于 2026 年 10 月 3 日] 标题：UniPro：通过传播实现从 2D 图像到 3D 体积的统一多模式医学图像分割。

Abstract: Medical image segmentation remains fragmented along two axes: segmentation paradigms and data dimensionality. Existing methods are typically developed separately for semantic, in-context, and interactive segmentation, and are further specialized to either native 2D images or 3D volumetric data.

摘要：医学图像分割在分割范式和数据维度两个轴向上仍然处于碎片化状态。现有的方法通常针对语义分割、上下文分割和交互式分割分别开发，并进一步专门用于原生 2D 图像或 3D 体积数据。

In clinical practice, however, segmentation workflows take many forms: a case may be initialized by semantic prediction, reference-guided segmentation, or user interaction. Regardless of how it begins, fine-grained refinement is naturally performed on 2D views; for volumetric scans, such 2D edits must propagate coherently to the rest of the volume.

然而，在临床实践中，分割工作流程有多种形式：病例可以通过语义预测、参考引导分割或用户交互来初始化。无论以何种方式开始，细粒度的精修通常是在 2D 视图上进行的；对于体积扫描，此类 2D 编辑必须连贯地传播到体积的其余部分。

We present UniPro, a unified model that bridges segmentation paradigms and data dimensionality, using propagation to extend 2D segmentation to 3D volumes. Our key insight is that volumetric propagation and in-context segmentation share the same reference-conditioned prediction mechanism, differing only in whether the reference image-mask pairs come from other cases or from previously segmented neighboring slices.

我们提出了 UniPro，这是一个连接分割范式和数据维度的统一模型，利用传播将 2D 分割扩展到 3D 体积。我们的核心见解是，体积传播和上下文分割共享相同的参考条件预测机制，区别仅在于参考图像-掩码对是来自其他病例还是来自先前已分割的相邻切片。

Building on this view, UniPro supports semantic, in-context, interactive, and propagation-based segmentation within a single slice-based framework, using class priors, reference exemplars, user clicks, and neighboring-slice predictions as mode-specific conditioning inputs.

基于这一观点，UniPro 在单一的基于切片的框架内支持语义、上下文、交互式和基于传播的分割，并使用类别先验、参考示例、用户点击和相邻切片预测作为特定模式的条件输入。

To improve propagation reliability, UniPro further incorporates bidirectional and 3D supervision to regularize slice-wise propagation beyond per-slice losses. Extensive experiments across diverse modalities and anatomies show that UniPro achieves strong performance across all segmentation settings, enabling annotation-efficient 3D segmentation from sparse 2D initialization and reducing slice-by-slice correction effort.

为了提高传播的可靠性，UniPro 进一步结合了双向和 3D 监督，以在逐切片损失之外规范化切片间的传播。在多种模态和解剖结构上的广泛实验表明，UniPro 在所有分割设置中均表现出色，实现了从稀疏 2D 初始化进行高效标注的 3D 分割，并减少了逐切片校正的工作量。