TEAMS：文本提示的时空双路头Mamba Snake
TEAMS: Text-prompted spatiotEmporal dual-heAd Mamba Snake
- School of intelligent systems engineering, Sun Yat-sen University（中山大学智能系统工程学院）
- Affiliated Hangzhou First People’s Hospital, Zhejiang University School of Medicine（浙江大学医学院附属杭州市第一人民医院）
- School of Engineering, Case Western Reserve University（凯斯西储大学工程学院）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
该研究提出视觉-语言Mamba框架TEAMS，通过三项创新改进深度蛇方法，在五个医学图像数据集上较现有方法取得显著性能提升，可作为多种医学图像分割场景的可靠工具。
中文摘要 AI 辅助
深度蛇（Deep snake）是一类极具潜力的实例分割方法，可准确预测物体级轮廓，从而克服语义分割方法中常见的像素级误分类问题，如掩码空洞和锯齿边缘。然而，现有深度蛇方法在处理复杂形态变化、准确捕捉细粒度器官细节以及修正基础检测误差方面面临挑战。为缓解这些局限，我们提出了一种 cohesive Text-prompted spatiotEmporal dual-heAd Mamba Snake（TEAMS），这是一种新型视觉-语言Mamba蛇框架，包含三项关键创新：（1）引入时空蛇演化策略（SSES），通过在状态空间模型中捕捉沿蛇轮廓的双向空间依赖关系及演化步骤间的时间动态，来应对复杂形态变化；（2）提出轮廓形态感知Mamba（CMAM），用于量化局部轮廓形态，以调制Mamba2 SSD双形式中的结构化注意力掩码，这扩展了Mamba感知输入序列元素相对重要性的能力，从而更好地勾勒细粒度器官细节；（3）设计文本提示协同双路头蛇（TCDHS），用于整合文本提示的线索并将演化后的轮廓信息传递至基础检测头，这增强了深度蛇的工作流程并减少错误检测。在涵盖不同器官和成像模态的五个数据集上进行的综合评估表明，TEAMS优于现有的语义分割和深度蛇分割方法（例如，在脊柱数据集中相对mDice/mBF提升6.9%/9.1%），凸显了其作为适用于多种医学图像分割场景的可靠工具的潜力。
英文摘要
Deep snake is a promising family of instance segmentation methods that accurately predicts object-level contours, thereby overcoming common pixel-level misclassification issues such as mask cavities and jagged edges in semantic segmentation approaches. However, existing deep snake methods face challenges in handling complex morphological variations, accurately capturing fine-grained organ details, and correcting base detection errors. To mitigate these limitations, we propose a cohesive Text-prompted spatiotEmporal dual-heAd Mamba Snake (TEAMS), a novel vision-language Mamba snake framework with three key innovations: (1) A Spatiotemporal Snake Evolution Strategy (SSES) is introduced to tackle complex morphological variations by capturing bidirectional spatial dependencies along the snake contour and temporal dynamics across evolution steps in a state space model. (2) A Contour Morphology-Aware Mamba (CMAM) is proposed to quantify local contour morphologies to modulate the structured attention mask in the Mamba2 SSD dual form, which extends Mamba's capability to perceive the relative importance of its input sequence elements for better delineation of fine-grained organ details. (3) A Text-prompted Collaborative Dual-Head Snake (TCDHS) is designed to incorporate cues from textual prompts and transfer the evolved contour information to the base detection head, which enhances the deep snake workflow and mitigates wrong detections. Comprehensive evaluations on five datasets covering different organs and imaging modalities demonstrate that TEAMS outperforms existing semantic and deep snake segmentation methods (e.g., relative mDice/mBF improvements of 6.9%/9.1% in a spinal dataset), underscoring its potential as a reliable tool across diverse medical image segmentation scenarios.
