TK-Mamba：将KAN与Mamba结合用于文本驱动的3D医学图像分割
TK-Mamba: Marrying KAN With Mamba for Text-Driven 3D Medical Image Segmentation
AI总结：
TK-Mamba通过融合KAN与Mamba，提出文本驱动的3D医学图像分割方法，实现高效准确的多器官和肿瘤分割。
摘要：
3D医学图像分割对于临床诊断和治疗至关重要，但面临高维数据和复杂空间依赖性的挑战。传统单模态网络，如CNN和Transformer，往往在三维设置中受限于计算效率低下和受限的上下文建模。为缓解这些限制，我们提出了TK-Mamba，一种多模态框架，将线性时间Mamba与Kolmogorov-Arnold网络（KAN）结合，形成一个高效的混合主干。我们的方法有两个主要技术贡献。首先，我们引入了新的3D-Group-Rational KAN（3D-GR-KAN），这是KAN在3D医学成像中的首次应用，提供了一种优越且计算高效的非线性特征变换，对于复杂的体积分割至关重要。其次，我们设计了一种双分支文本驱动策略，利用Pubmedclip的嵌入。该策略通过同时捕捉器官间的语义关系来增强分割的鲁棒性和准确性，以缓解标签不一致并使图像特征与解剖文本对齐。通过结合这种先进的主干和视觉-语言知识，TK-Mamba为多器官和肿瘤分割提供了一个统一且可扩展的解决方案。在多个数据集上的实验表明，我们的框架在器官和肿瘤分割任务中均实现了最先进的性能，优于现有方法在准确性和效率方面。我们的代码已公开在https://github.com/yhy-whu/TK-Mamba。
英文摘要：
3D medical image segmentation is important for clinical diagnosis and treatment but faces challenges from high-dimensional data and complex spatial dependencies. Traditional single-modality networks, such as CNNs and Transformers, are often limited by computational inefficiency and constrained contextual modeling in 3D settings. To alleviate these limitations, we propose TK-Mamba, a multimodal framework that fuses the linear-time Mamba with Kolmogorov-Arnold Networks (KAN) to form an efficient hybrid backbone. Our approach is characterized by two primary technical contributions. Firstly, we introduce the novel 3D-Group-Rational KAN (3D-GR-KAN), which marks the first application of KAN in 3D medical imaging, providing a superior and computationally efficient nonlinear feature transformation crucial for complex volumetric structures. Secondly, we devise a dual-branch text-driven strategy using Pubmedclip's embeddings. This strategy significantly enhances segmentation robustness and accuracy by simultaneously capturing inter-organ semantic relationships to mitigate label inconsistencies and aligning image features with anatomical texts. By combining this advanced backbone and vision-language knowledge, TK-Mamba offers a unified and scalable solution for both multi-organ and tumor segmentation. Experiments on multiple datasets demonstrate that our framework achieves state-of-the-art performance in both organ and tumor segmentation tasks, surpassing existing methods in both accuracy and efficiency. Our code is publicly available at https://github.com/yhy-whu/TK-Mamba
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
