Automated segmentation and quantification of geometrical and structural parameters of the hip in 3D-MRI data sets: Development of algorithms for reliable data analysis in large cohorts (German National Cohort)

德国国家队列（GNC）成像子研究约30，000例受试者的MRI数据评估需要高度自动化的标准化和验证工具。本项目的目的是开发和测试软件算法，用于利用GNC提供的200个示例性MRI数据集的骨盆序列进行髋关节自动分割。软件算法将自动量化以下成像参数：髋臼覆盖、股骨头球形度、偏心距股骨头/颈、股骨近端形状、软骨面积和厚度、髋臼和股骨囊肿。将这些自动导出的参数与手动分析进行比较，并在单独的临床相关患者队列中进行验证。最后，开发的软件工具将允许自动定量评估结构变化，这可能使个人发展髋关节骨关节炎。

Standardized and validated tools with a high degree of automatization are required for the assessment of the MRI-data of the approximately 30,000 participants of the German National Cohort (GNC) imaging sub-study. The purpose of this project is to develop and test software algorithms for automated segmentation of the hip joint exploiting the pelvis sequences of 200 exemplary MRI datasets provided by the GNC. The software algorithms will automatically quantify the following imaging parameters: Acetabular coverage, sphericity of the femoral head, offset femoral head/neck, proximal femoral form, cartilage area and thickness, acetabular and femoral cysts. These automatically derived parameters will be compared to manual analysis and validated in a separate clinically correlated patient cohort. Finally, the developed software tools will allow for automated quantitative assessment of structural changes which may predispose individuals to develop hip osteoarthritis.