GraPHFormer：用于神经科学形态分析的多模态图持续同调 Transformer
GraPHFormer: A Multimodal Graph Persistent Homology Transformer for the Analysis of Neuroscience Morphologies
把持续同调得到的持续图像当作三通道视觉输入、用 DINOv2-ViT 提取特征，再与 TreeLSTM 的骨架图特征做 CLIP 式对称 InfoNCE 对齐——这套「拓扑图像 + 图结构」双分支对比框架可直接迁移到其他需要同时建模形状拓扑与骨架几何的视觉任务（如血管、道路、细胞骨架分析）。
原文依据 Our vision branch processes a novel three-channel persistence image encoding unweighted, persistence-weighted, and radius-weighted topological densities via DINOv2-ViT-S.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论神经元形态编码了关于回路功能、发育与疾病的关键信息，然而现有方法往往孤立地分析拓扑结构或图结构。我们提出 GraPHFormer，一种通过 CLIP 式对比学习将这两种互补视角统一起来的多模态架构。其视觉分支通过 DINOv2-ViT-S 处理一种新颖的三通道持续图像，该图像编码了无权、持续加权与半径加权的拓扑密度。与此同时，一个 TreeLSTM 编码器从骨架图中捕获几何与径向属性。两个分支被投影到共享嵌入空间，并以对称 InfoNCE 损失进行训练，同时辅以保持拓扑语义的持续空间变换。在涵盖自监督与监督设置的六个基准（BIL-6、ACT-4、JML-4、N7、M1-Cell、M1-REG）上评估，GraPHFormer 在其中五个基准上取得最优性能，显著优于仅拓扑、仅图以及形态计量学基线。我们通过区分不同皮层区域与物种间的胶质细胞形态，以及检测发育与退行过程的特征，展示了其实际效用。代码：https://github.com/Uzshah/GraPHFormer
英文原文摘要 1224 字符
Neuronal morphology encodes critical information about circuit function, development, and disease, yet current methods analyze topology or graph structure in isolation. We introduce GraPHFormer, a multimodal architecture that unifies these complementary views through CLIP-style contrastive learning. Our vision branch processes a novel three-channel persistence image encoding unweighted, persistence-weighted, and radius-weighted topological densities via DINOv2-ViT-S. In parallel, a TreeLSTM encoder captures geometric and radial attributes from skeleton graphs. Both project to a shared embedding space trained with symmetric InfoNCE loss, augmented by persistence-space transformations that preserve topological semantics. Evaluated on six benchmarks (BIL-6, ACT-4, JML-4, N7, M1-Cell, M1-REG) spanning self-supervised and supervised settings, GraPHFormer achieves state-of-the-art performance on five benchmarks, significantly outperforming topology-only, graph-only, and morphometrics baselines. We demonstrate practical utility by discriminating glial morphologies across cortical regions and species, and detecting signatures of developmental and degenerative processes. Code: https://github.com/Uzshah/GraPHFormer