What does this research mean for the field?

ZR2ViM achieves a 2.15 mm reduction in the HD95 on the Synapse multi-organ CT dataset compared to the CC-ViM baseline, demonstrating superior boundary-preserving segmentation capabilities across diverse medical imaging modalities. Novelty: ClaimNovelty.NOVEL_FINDING. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To enhance medical image segmentation by developing a recursive model that maintains boundary precision while managing computational resources.

March 7, 2026Open Access

ZR2ViM: a recursive vision Mamba model for boundary-preserving medical image segmentation

Puntos clave

To enhance medical image segmentation by developing a recursive model that maintains boundary precision while managing computational resources.
Developed ZR2ViM model using Zigzag Recursive Reinforced Block.
Incorporated Stacked State Redistribution and Nested Recursive Connection for detail and context fusion.
Executed multi-step recursive updates with Cross-directional Zigzag WKV module.
Evaluated model across four medical imaging domains and five public datasets.
ZR2ViM outperformed convolutional and attention-based models in segmentation accuracy.
Achieved a 2.15 mm reduction in HD95 metric on the Synapse multi-organ CT dataset compared to the baseline model.
Demonstrated consistent performance improvements in region consistency and boundary localization across all datasets.

Resumen

Introduction Medical image segmentation is fundamental to quantitative disease analysis and therapeutic decision-making. However, constrained by limited computational resources, existing deep learning methods often struggle to simultaneously model long-range dependencies and preserve boundary precision, particularly when delineating structures with complex morphology or blurred edges. Method To overcome these challenges, we propose ZR 2 ViM, a recursion-enhanced visual state space model designed for medical image segmentation. ZR 2 ViM augments the Vision Mamba framework with a Zigzag Recursive Reinforced ( ZR 2 ) Block that incorporates Stacked State Redistribution (SSR) and a Nested Recursive Connection (NRC). The NRC employs dual inner and outer pathways to iteratively fuse local details with global context while preserving 2D spatial adjacency. Furthermore, a Cross-directional Zigzag WKV (CZ-WKV) module executes multi-step recursive updates along multiple zigzag trajectories, injecting spatial directional information via Quad-Directional Token Shift (Q-Shift) directional priors. Collectively, these mechanisms mitigate serialization-induced banding artifacts and enhance the representation of fine, elongated, and low-contrast structures, all while maintaining near-linear computational complexity. Results Comprehensive evaluations across four medical imaging domains—spanning dermatoscopic images, breast ultrasound, colorectal polyps, and abdominal multi-organ CT—on five public datasets demonstrate that ZR 2 ViM consistently outperforms representative convolutional, attention-based, and visual state space architectures in region consistency and boundary localization. Notably, ZR 2 ViM achieves a 2.15 mm reduction in the HD95 on the Synapse multi-organ CT dataset relative to the CC-ViM baseline, substantiating its superior capability for precise, clinically relevant boundary delineation. Conclusion The ZR 2 ViM framework delivers accurate, boundary-preserving segmentation across diverse imaging modalities and anatomically complex structures, achieving these gains with near-linear computational complexity. These findings demonstrate that ZR 2 ViM offers a robust and efficient solution for medical image analysis, establishing a promising foundation for advanced clinical and research applications.

Me gusta

Guardar

Ver artículo completo