Spinal Cord Segmentation Model
收藏资源简介:
本数据集是一项针对脊髓分割的模型研究,旨在用于监测脊髓形态学变化,如多发性硬化症中的脊髓萎缩和退行性颈椎脊髓病的压缩程度。数据集由来自全球75个地点的1,631名参与者提供,包括健康参与者、退行性颈椎脊髓病患者、脊髓损伤患者、多发性硬化症患者、肌萎缩侧索硬化症患者、视神经脊髓炎患者和脊髓空洞症患者。数据集包含9种不同的MRI对比度,跨越广泛的图像分辨率,并包括多种脊髓病理学。该模型通过多站数据集训练,并引入了终身学习框架,以自动监测模型更新时的形态学漂移。该模型已在Spinal Cord Toolbox v7.0中免费提供。
This dataset is a model study targeting spinal cord segmentation, intended for monitoring morphological changes of the spinal cord, such as spinal cord atrophy in multiple sclerosis and the degree of compression in degenerative cervical myelopathy. The dataset is contributed by 1,631 participants from 75 global sites, including healthy volunteers, patients with degenerative cervical myelopathy, spinal cord injury, multiple sclerosis, amyotrophic lateral sclerosis, neuromyelitis optica, and syringomyelia. It encompasses 9 distinct MRI contrasts, spans a wide range of image resolutions, and covers multiple spinal cord pathologies. The model is trained on this multi-site dataset and incorporates a lifelong learning framework to automatically monitor morphological drift during model updates. This model is freely available in Spinal Cord Toolbox v7.0.

- 1Monitoring morphometric drift in lifelong learning segmentation of the spinal cordNeuroPoly Lab, Institute of Biomedical Engineering, Polytechnique Montreal, Montreal, QC, Canada · 2025年



