Servier Virtual Cardiac Centre and Department of Radiology and Diagnostic Imaging, University of Alberta, Edmonton, T6G 2B7, Canada.
Int J Comput Assist Radiol Surg. 2017 Mar;12(3):439-447. doi: 10.1007/s11548-016-1510-4. Epub 2016 Dec 26.
Developmental dysplasia of the hip (DDH) is a congenital deformity which in severe cases leads to hip dislocation and in milder cases to premature osteoarthritis. Image-aided diagnosis of DDH is partly based on Graf classification which quantifies the acetabular shape seen at two-dimensional ultrasound (2DUS), which leads to high inter-scan variance. 3D ultrasound (3DUS) is a promising alternative for more reliable DDH diagnosis. However, manual quantification of acetabular shape from 3DUS is cumbersome.
Here, we (1) propose a semiautomated segmentation algorithm to delineate 3D acetabular surface models from 3DUS using graph search; (2) propose a fully automated method to classify acetabular shape based on a random forest (RF) classifier using features derived from 3D acetabular surface models; and (3) test diagnostic accuracy on a dataset of 79 3DUS infant hip recordings (36 normal, 16 borderline, 27 dysplastic based on orthopedic surgeon assessment) in 42 patients. For each 3DUS, we performed semiautomated segmentation to produce 3D acetabular surface models and then calculated geometric features including the automatic [Formula: see text]lpha (AA), acetabular contact angle (ACA), kurtosis (K), skewness (S) and convexity (C). Mean values of features obtained from surface models were used as inputs to train a RF classifier.
Surface models were generated rapidly (user time 46.2 s) via semiautomated segmentation and visually closely correlated with actual acetabular contours (RMS error 1.39 ± 0.7 mm). A paired nonparametric u test on of feature values in each category showed statistically significant variation (p < 0.001) for AA, ACA and convexity. The RF classifier was 100 % specific and 97.2 % sensitive in classifying normal versus dysplastic hips and yielded true positive rates of 94.4, 62.5 and 89.9 % for normal, borderline and dysplastic hips.
The proposed technique reduces the subjectivity of image-aided DDH diagnosis and could be useful in clinical practice.
发育性髋关节发育不良（DDH）是一种先天性畸形，在严重的情况下会导致髋关节脱位，在轻微的情况下会导致早期骨关节炎。DDH 的影像学诊断部分基于 Graf 分类，该分类量化了二维超声（2DUS）所见的髋臼形态，导致扫描间变异性高。三维超声（3DUS）是一种更可靠的 DDH 诊断的有前途的替代方法。然而，从 3DUS 手动量化髋臼形态很繁琐。
在这里，我们（1）提出了一种使用图搜索半自动分割算法从 3DUS 中描绘 3D 髋臼表面模型；（2）提出了一种完全自动化的方法，使用从 3D 髋臼表面模型中提取的特征，基于随机森林（RF）分类器对髋臼形状进行分类；（3）在 42 名患者的 79 个 3DUS 婴儿髋关节记录（36 个正常，16 个边界，27 个发育不良，根据矫形外科医生评估）的数据集上测试诊断准确性。对于每个 3DUS，我们都进行了半自动分割以生成 3D 髋臼表面模型，然后计算了包括自动 [公式：见文本]lpha（AA）、髋臼接触角（ACA）、峰度（K）、偏度（S）和凸度（C）在内的几何特征。从表面模型获得的特征的平均值用作输入来训练 RF 分类器。
通过半自动分割快速生成表面模型（用户时间 46.2s），并且与实际髋臼轮廓视觉上密切相关（均方根误差 1.39±0.7mm）。对每个类别的特征值进行的配对非参数 u 检验显示 AA、ACA 和凸度的统计学显着变化（p<0.001）。RF 分类器在正常与发育不良髋关节之间的分类中具有 100%的特异性和 97.2%的敏感性，对于正常、边界和发育不良髋关节，其真阳性率分别为 94.4%、62.5%和 89.9%。
所提出的技术减少了影像学辅助 DDH 诊断的主观性，并且在临床实践中可能有用。
