This paper presents a large publicly available multi-center lumbar spine magnetic resonance imaging (MRI) dataset with reference segmentations of vertebrae, intervertebral discs (IVDs), and spinal canal. The dataset includes 447 sagittal T1 and T2 MRI series from 218 patients with a history of low back pain. It was collected from four different hospitals and was divided into a training (179 patients) and validation (39 patients) set. An iterative data annotation approach was used by training a segmentation algorithm on a small part of the dataset, enabling semi-automatic segmentation of the remaining images. The algorithm provided an initial segmentation, which was subsequently reviewed, manually corrected, and added to the training data. We provide reference performance values for this baseline algorithm and nnU-Net, which performed comparably. We set up a continuous segmentation challenge to allow for a fair comparison of different segmentation algorithms. This study may encourage wider collaboration in the field of spine segmentation, and improve the diagnostic value of lumbar spine MRI.

翻译：本文介绍了一个大规模公开的多中心腰椎磁共振成像（MRI）数据集，其中包含椎体、椎间盘（IVDs）和椎管的参考分割结果。该数据集包含来自218名有下背痛病史患者的447个矢状位T1和T2 MRI序列，采集自四家不同医院，并划分为训练集（179名患者）和验证集（39名患者）。我们采用迭代式数据标注方法：先在数据集的一小部分上训练分割算法，使其能够对其他图像进行半自动分割。该算法提供初始分割结果，随后经人工复核、手动修正后加入训练数据。我们提供了该基线算法及表现相当的nnU-Net的参考性能值。为便于不同分割算法的公平比较，我们设立了持续性的分割挑战赛。本研究有望促进脊柱分割领域的更广泛合作，并提升腰椎MRI的诊断价值。