发表机构
University of Massachusetts Boston; Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School; Universidad San Sebastián; Université Claude Bernard Lyon 1, Université de Lyon; Columbia University in the City of New York; University College London（马萨诸塞大学波士顿分校; 阿西诺拉·A·马蒂诺斯生物医学成像中心，马萨诸塞总医院和哈佛医学院; 圣塞巴斯蒂安大学; 里昂第一大学，里昂大学; 纽约市哥伦比亚大学; 伦敦大学学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出AxonSynth域随机化合成数据框架，训练3D U-Net实现零样本轴突分割，无需人工标注，在猕猴和人类光片显微镜数据上优于传统方法，显著降低组件计数误差。
AI 中文摘要
在3D显微镜数据中准确分割轴突对于分析白质组织至关重要，但密集的地面真值标签获取成本高昂。现有的监督式轴突分割方法依赖目标域注释，当组织类型、物种、模态或采集条件变化时可能变得脆弱。我们提出AxonSynth，一个域随机化合成数据框架，用于训练3D轴突分割模型，无需手动标注的真实训练体积。AxonSynth生成带有方向先验的密集合成轴突标签，反映真实的纤维配置，并以随机密度、对比度、偏置场、模糊和噪声进行渲染。训练一个三类别3D U-Net来预测背景、轴突鞘和轴突内空间。我们在来自猕猴和人类大脑样本的10个保留光片显微镜（LSM）块上评估零样本迁移，这些块使用三种轴突标记之一进行标记，并与校准阈值法和Frangi滤波法在重叠、校正检测、假阳性和拓扑指标上进行比较。在猕猴样本上，AxonSynth实现了最佳的校正Dice和校正精度（0.826和0.851），而阈值法为0.765和0.754，Frangi法为0.685和0.762。在人类样本上，校正Dice与阈值法相当（0.857对比0.868），而组件计数误差从22,504降至3,377。在所有保留块中，AxonSynth在10/10块中减少了组件计数误差，在8/10块中减少了欧拉特征误差。这些结果表明，合成标签域随机化可以减少对人工轴突标注的依赖，同时支持合成到真实的3D分割。
英文摘要
Accurate segmentation of axons in 3D microscopy data is important for analyzing white-matter organization, but dense ground truth labels are expensive to obtain. Existing supervised axon segmentation methods rely on target-domain annotations and can be brittle when tissue type, species, modality, or acquisition conditions change. We present AxonSynth, a domain-randomized synthetic-data framework for training 3D axon segmentation models without manually annotated real training volumes. AxonSynth generates dense synthetic axon labels with orientation priors that reflect realistic fiber configurations and renders them with randomized density, contrast, bias fields, blur, and noise. A three-class 3D U-Net is trained to predict background, axon sheath and intra-axonal space. We evaluate zero-shot transfer on 10 held-out light-sheet microscopy (LSM) patches from macaque and human brain samples labeled with one of three axonal markers, comparing against calibrated thresholding and Frangi filtering using overlap, corrected detection, false-positive, and topology metrics. On macaque samples, AxonSynth achieved the best corrected Dice and corrected precision (0.826 and 0.851), compared with 0.765 and 0.754 for thresholding and 0.685 and 0.762 for Frangi. On human samples, corrected Dice was comparable to thresholding (0.857 vs. 0.868), while component-count error decreased from 22,504 to 3,377. Across all held-out patches, AxonSynth reduced component-count error in 10/10 patches and Euler-characteristic error in 8/10. These results show that synthetic-label domain randomization can reduce dependence on manual axon annotation while supporting synthetic-to-real 3D segmentation.
Comments11 pages, 2 figures, 2 tables. Accepted at SASHIMI 2026, held with MICCAI 2026
