CoInS-Net: A Continuous Position-Aware Network for Joint Medical Image Interpolation and Segmentation

2026-08-10

AI Summary

This study addresses the challenges of structural discontinuity and blurred boundaries in anisotropic medical images caused by sparse inter-slice sampling. Existing approaches typically handle interpolation and segmentation as separate tasks, failing to exploit cross-slice structural information effectively. To overcome this limitation, the authors propose CoInS-Net, a novel joint learning framework that integrates continuous positional awareness. CoInS-Net employs a shared Swin Transformer encoder, augmented with a continuous spatial coordinate query mechanism, a prototype-driven task interaction module, and a multi-scale collaborative decoder, enabling bidirectional optimization between interpolation and segmentation without requiring additional annotations. Extensive experiments on four multimodal public datasets demonstrate that the proposed method significantly outperforms single-task baselines, confirming the mutual benefits between tasks and its clinical applicability.

Abstract

Accurate medical image interpolation and anatomical structure segmentation are fundamental for computer-aided diagnosis and treatment planning. Anisotropic medical volumes with sparse through-plane sampling often suffer from structural discontinuity and boundary blur, hindering reliable clinical image analysis. Most existing methods implement interpolation and segmentation independently, which introduces redundant computation and fails to fully exploit complementary cross-slice structural information between sequential slices. To address these issues, we propose a continuous position-aware interaction network, termed CoInS-Net, for joint frame interpolation and lesion segmentation. Unlike conventional cascaded interpolation-then-segmentation paradigms, the framework enables bidirectional interaction under a shared Swin encoder with continuous spatial coordinate queries. A spatially continuous position interpolation module generates target-position features at every scale from the relative coordinate and physical spacing, and a prototype-based task mutual interaction module lets the segmentation and interpolation branches exchange global structure through a small set of shared prototypes rather than dense feature mixing. A multi-scale task-cooperative decoder further separates each scale into shared and task-specific components, so the two tasks reinforce common anatomy while preserving their distinct requirements down to the boundary level, without extra annotations. Experiments on four public medical imaging datasets with diverse modalities and anatomical regions demonstrate that the proposed method outperforms conventional single-task schemes. The joint optimization framework effectively realizes mutual promotion between interpolation and segmentation tasks, providing a reliable and universal technical scheme for intelligent clinical medical image analysis.

Problem

• medical image interpolation
• anatomical segmentation
• anisotropic volumes
• structural discontinuity
• cross-slice information

Innovation

• joint interpolation and segmentation
• continuous position-aware
• prototype-based interaction
• multi-scale task cooperation
• medical image analysis