FiberGeoText: A Vision-Language Model for Population- Level Organization of Superficial White Matter

📄 arXiv: 2610.02755v1 📥 PDF

作者: Yuqian Chen, R. Jarrett Rushmore, Guikun Chen, Fan Zhang, Edward Yeterian, Nikos Makris, Yogesh Rathi, Lauren J. O'Donnell

分类: cs.CV

发布日期: 2026-10-02

备注: 22 pages, 3 figures

💡 一句话要点

提出FiberGeoText模型以解决白质组织的个体差异问题

🎯 匹配领域: 支柱九：具身大模型 (Embodied Foundation Models)

关键词: 浅层白质 视觉-语言模型 群体级聚类 多模态深度嵌入 医学影像分析 神经科学 个体差异

📋 核心要点

  1. 现有方法在组织短程白质纤维时面临个体间解剖差异导致的对应困难，影响了对其空间位置的准确描述。
  2. 本文提出FiberGeoText模型，通过联合表示流线的几何、解剖和形状信息，克服了个体差异带来的挑战。
  3. 实验结果显示，FGT在聚类一致性和跨个体对应性方面显著优于现有最先进方法，且在未见样本上恢复率高达96.7%。

📝 摘要（中文）

浅层白质(SWM)是大脑中一个关键区域，涉及认知和脑部疾病，其短程关联纤维的组织尚未完全表征。由于个体间的皮层折叠差异，传统方法难以准确对应这些纤维。本文提出FiberGeoText(FGT)，一种视觉-语言模型(VLM)，用于将超高分辨率扩散MRI重建的SWM流线组织为群体级聚类。FGT联合表示每条流线的三维轨迹、皮层解剖背景和形状，利用预训练的大型语言模型对多种分区方案的皮层端点信息进行编码。实验结果表明，FGT在皮层包络一致性、聚类形状一致性等方面显著优于现有方法，并在未见样本上表现出良好的泛化能力。

🔬 方法详解

问题定义：本文旨在解决浅层白质(SWM)短程纤维的组织问题，现有方法因个体间解剖差异而难以准确对应这些纤维的空间位置。

核心思路：提出FiberGeoText(FGT)模型，通过联合表示流线的三维轨迹、皮层解剖背景和形状信息，利用视觉-语言模型(VLM)进行群体级聚类，从而克服个体差异带来的挑战。

技术框架：FGT的整体架构包括三个主要模块：流线的三维轨迹表示、皮层解剖上下文的文本编码，以及流线形状的特征提取。这些模块共同作用，形成一个统一的多模态深度嵌入表示。

关键创新：FGT的创新在于将几何、解剖和形状信息进行整合，通过学习多模态深度嵌入，显著提高了SWM组织的群体一致性，克服了传统方法的局限性。

关键设计：模型使用预训练的大型语言模型对皮层端点信息进行编码，采用适当的损失函数以优化聚类效果，确保聚类的一致性和形状的稳定性。

🖼️ 关键图片

img_0
img_1
img_2

📊 实验亮点

实验结果表明，FiberGeoText模型在皮层包络一致性、聚类形状一致性和跨个体对应性方面均显著优于现有最先进方法，聚类恢复率高达96.7%，展示了其在处理个体间解剖变异时的强大能力。

🎯 应用场景

该研究具有广泛的应用潜力，尤其在神经科学和医学影像分析领域。通过更准确地组织和理解浅层白质的结构，能够为脑部疾病的诊断和治疗提供更有力的支持，推动个性化医疗的发展。

📄 摘要（原文）

The superficial white matter (SWM), a critical brain region for cognition across the lifespan and brain disease, contains abundant short-range association fibers whose organization remains incompletely characterized, in part because the short trajectories and highly variable cortical folding make correspondence across individuals challenging. Anatomically corresponding connections may vary in spatial location across individuals and therefore may not be adequately defined by geometric proximity alone. We introduce FiberGeoText (FGT), a vision-language model (VLM) for organizing short-range superficial white matter (SWM) streamlines reconstructed from ultra-high-resolution diffusion MRI into population-level clusters. FGT jointly represents three complementary properties of each streamline: its three-dimensional trajectory, its cortical anatomical context, and its shape. Cortical endpoint information from multiple parcellation schemes is expressed as text and encoded using a pretrained large language model (LLM), enabling heterogeneous anatomical descriptions to contribute to a common continuous representation. We evaluated FGT on acquired submillimeter 0.76 mm diffusion MRI data. Compared with state-of-the-art (SOTA) methods, FGT produced substantially greater cortical parcel coherence, within-cluster shape consistency, cluster-size consistency, and cross-subject correspondence. The trained model also generalizes well to unseen subjects with an average of 96.7% of the 5,000 learned clusters recovered, and high consistency of cluster structure between training and testing data. Together, these findings demonstrate that integrating geometric, anatomical, and shape information by learning multimodal deep embeddings with a VLM model enables robust learning of population-consistent SWM organization despite interindividual anatomical variability.