Fischer Marc, Walter Sven S, Hepp Tobias, Zimmer Manuela, Notohamiprodjo Mike, Schick Fritz, Yang Bin
Institute of Signal Processing and Systems Theory, University of Stuttgart, Pfaffenwaldring 47, 70550 Stuttgart, Germany (M.F., M.Z., B.Y.); Department of Diagnostic and Interventional Radiology, Section on Experimental Radiology, University Hospital Tübingen, Tübingen, Germany (S.S.W., T.H., M.N., F.S.); and Empirical Inference Department, Max Planck Institute for Intelligent Systems, Tübingen, Germany (T.H.).
Radiol Artif Intell. 2021 Jun 2;3(5):e200213. doi: 10.1148/ryai.2021200213. eCollection 2021 Sep.
To develop and validate an automated morphometric analysis framework for the quantitative analysis of geometric hip joint parameters in MR images from the German National Cohort (GNC) study.
A secondary analysis on 40 participants (mean age, 51 years; age range, 30-67 years; 25 women) from the prospective GNC MRI study (2015-2016) was performed. Based on a proton density-weighted three-dimensional fast spin-echo sequence, a morphometric analysis approach was developed, including deep learning-based landmark localization, bone segmentation of the femora and pelvis, and a shape model for annotation transfer. The centrum-collum-diaphyseal, center-edge (CE), three alpha angles, head-neck offset (HNO), and HNO ratio along with the acetabular depth, inclination, and anteversion were derived. Quantitative validation was provided by comparison with average manual assessments of radiologists in a cross-validation format. Paired-sample tests with a Bonferroni-corrected significance level of .005 were employed alongside mean differences and 10th/90th percentiles, median absolute deviations (MADs), and intraclass correlation coefficients (ICCs).
High agreement in mean Dice similarity coefficients was achieved (average of 97.52% ± 0.46 [standard deviation]). The subsequent morphometric analysis produced results with low mean MAD values, with the highest values of 3.34° (alpha 03:00 o'clock position) and 0.87 mm (HNO) and ICC values ranging between 0.288 (HNO ratio) and 0.858 (CE) compared with manual assessments. These values were in line with interreader agreements, which at most had MAD values of 4.02° (alpha 12:00 o'clock position) and 1.07 mm (HNO) and ICC values ranging between 0.218 (HNO ratio) and 0.777 (CE).
Automatic extraction of geometric hip parameters from MRI is feasible using a morphometric analysis approach with deep learning. Computer-Aided Diagnosis (CAD), Interventional-MSK, MR-Imaging, Neural Networks, Skeletal-Appendicular, Hip, Anatomy, Computer Applications-3D, Segmentation, Vision, Application Domain, Quantification © RSNA, 2021.
开发并验证一种自动化形态计量分析框架，用于对德国国民队列（GNC）研究中的磁共振成像（MR）图像进行髋关节几何参数的定量分析。
对前瞻性GNC MRI研究（2015 - 2016年）中的40名参与者（平均年龄51岁；年龄范围30 - 67岁；25名女性）进行二次分析。基于质子密度加权三维快速自旋回波序列，开发了一种形态计量分析方法，包括基于深度学习的地标定位、股骨和骨盆的骨分割以及用于注释转移的形状模型。得出了股骨颈干、中心边缘（CE）、三个α角、头颈偏移（HNO）和HNO比率以及髋臼深度、倾斜度和前倾角。通过以交叉验证形式与放射科医生的平均手动评估进行比较，提供了定量验证。采用配对样本检验，其Bonferroni校正显著性水平为0.005，并结合均值差异、第10/90百分位数、中位数绝对偏差（MAD）和组内相关系数（ICC）。
在平均Dice相似系数方面达成了高度一致（平均值为97.52% ± 0.46 [标准差]）。随后的形态计量分析得出的结果具有较低的平均MAD值，与手动评估相比，最高值为3.34°（α 03:00位置）和0.87 mm（HNO），ICC值在0.288（HNO比率）至0.858（CE）之间。这些值与阅片者间的一致性相符，阅片者间一致性的MAD值最高为4.02°（α 12:00位置）和1.07 mm（HNO），ICC值在0.218（HNO比率）至0.777（CE）之间。
使用具有深度学习的形态计量分析方法从MRI自动提取髋关节几何参数是可行的。计算机辅助诊断（CAD）、介入肌肉骨骼（Interventional - MSK）、磁共振成像（MR - Imaging）、神经网络、骨骼附属结构、髋关节、解剖学、计算机应用 - 3D、分割、视觉、应用领域、量化 © RSNA，2021年
