Birnbaum Andrew M, Buchwald Adam, Turkeltaub Peter, Jacks Adam, Carr George, Kannan Shreya, Huang Yu, Datta Abhisheck, Parra Lucas C, Hirsch Lukas A
The City College of New York, Department of Biomedical Engineering, New York, NY, USA.
New York University, New York, NY, USA.
ArXiv. 2025 Sep 2:arXiv:2501.18716v2.
The goal of this work was to develop a deep network for whole-head segmentation including clinical MRIs with abnormal anatomy, and compile the first public benchmark dataset for this purpose. We collected 98 MRIs with volumetric segmentation labels for a diverse set of human subjects including normal, as well as abnormal anatomy in clinical cases of stroke and disorders of consciousness.
Training labels were generated by manually correcting initial automated segmentations for skin/scalp, skull, CSF, gray matter, white matter, air cavity and extracephalic air. We developed a "MultiAxial" network consisting of three 2D U-Net that operate independently in sagittal, axial and coronal planes and are then combined to produce a single 3D segmentation.
The MultiAxial network achieved a test-set Dice scores of 0.88±0.04 (median ± interquartile range) on whole head segmentation including gray and white matter. This compared to 0.86 ± 0.04 for Multipriors and 0.79 ± 0.10 for SPM12, two standard tools currently available for this task. The MultiAxial network gains in robustness by avoiding the need for coregistration with an atlas. It performed well in regions with abnormal anatomy and on images that have been de-identified. It enables more accurate and robust current flow modeling when incorporated into ROAST, a widely-used modeling toolbox for transcranial electric stimulation.
We are releasing a new state-of-the-art tool for whole-head MRI segmentation in abnormal anatomy, along with the largest volume of labeled clinical head MRIs including labels for non-brain structures. Together the model and data may serve as a benchmark for future efforts.
本研究的目标是开发一个用于全脑分割的深度网络，该网络要能处理包含异常解剖结构的临床磁共振成像（MRI），并为此编制首个公开的基准数据集。我们收集了98例带有体积分割标签的MRI数据，涵盖了包括正常人群以及中风和意识障碍临床病例中的异常解剖结构的各类人类受试者。
通过手动校正皮肤/头皮、颅骨、脑脊液、灰质、白质、气腔和颅外空气的初始自动分割结果来生成训练标签。我们开发了一个“多轴”网络，它由三个在矢状面、轴面和冠状面独立运行的二维U-Net组成，然后将其组合以生成单个三维分割结果。
多轴网络在包括灰质和白质的全脑分割上，在测试集上的骰子系数得分为0.88±0.04（中位数±四分位间距）。相比之下，目前用于此任务的两个标准工具，多先验模型（Multipriors）的得分为0.86±0.04，统计参数映射（SPM12）的得分为0.79±0.10。多轴网络通过避免与图谱进行配准而增强了鲁棒性。它在具有异常解剖结构的区域以及已去识别的图像上表现良好。当将其整合到用于经颅电刺激的广泛使用的建模工具箱ROAST中时，它能实现更准确和稳健的电流建模。
我们正在发布一种用于异常解剖结构的全脑MRI分割的新的先进工具，以及包含非脑结构标签的最大规模的标记临床头部MRI数据集。模型和数据一起可作为未来研究的基准。
