Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland.
Department of Radiology & Nuclear Medicine, Erasmus MC University Medical Center, Rotterdam, the Netherlands.
Osteoporos Int. 2022 Feb;33(2):355-365. doi: 10.1007/s00198-021-06130-y. Epub 2021 Sep 2.
We developed and compared deep learning models to detect hip osteoarthritis on clinical CT. The CT-based summation images, CT-AP, that resemble X-ray radiographs can detect radiographic hip osteoarthritis and in the absence of large training data, a reliable deep learning model can be optimized by combining CT-AP and X-ray images.
In this study, we aimed to investigate the applicability of deep learning (DL) to assess radiographic hip osteoarthritis (rHOA) on computed tomography (CT).
The study data consisted of 94 abdominopelvic clinical CTs and 5659 hip X-ray images collected from Cohort Hip and Cohort Knee (CHECK). The CT slices were sequentially summed to create radiograph-like 2-D images named CT-AP. X-ray and CT-AP images were classified as rHOA if they had osteoarthritic changes corresponding to Kellgren-Lawrence grade 2 or higher. The study data was split into 55% training, 30% validation, and 15% test sets. A pretrained ResNet18 was optimized for a classification task of rHOA vs. no-rHOA. Five models were trained using (1) X-rays, (2) downsampled X-rays, (3) combination of CT-AP and X-ray images, (4) combination of CT-AP and downsampled X-ray images, and (5) CT-AP images.
Amongst the five models, Model-3 and Model-5 performed best in detecting rHOA from the CT-AP images. Model-3 detected rHOA on the test set of CT-AP images with a balanced accuracy of 82.2% and was able to discriminate rHOA from no-rHOA with an area under the receiver operating characteristic curve (ROC AUC) of 0.93 [0.75-0.99]. Model-5 detected rHOA on the test set at a balanced accuracy of 82.2% and classified rHOA from no-rHOA with an ROC AUC of 0.89 [0.67-0.97].
CT-based summation images that resemble radiographs can be used to detect rHOA. In addition, in the absence of large training data, a reliable DL model can be optimized by combining CT-AP and X-ray images.
本研究旨在探讨深度学习（DL）在评估计算机断层扫描（CT）中髋关节骨关节炎（rHOA）的适用性。
研究数据包括来自队列髋关节和队列膝关节（CHECK）的 94 例腹盆腔临床 CT 和 5659 例髋关节 X 射线图像。通过连续对 CT 切片进行求和，创建类似于 X 射线的 2-D 图像，命名为 CT-AP。如果 X 射线或 CT-AP 图像具有与 Kellgren-Lawrence 分级 2 或更高的对应改变，则将其分类为 rHOA。研究数据分为 55%的训练集、30%的验证集和 15%的测试集。使用（1）X 射线、（2）X 射线下采样、（3）CT-AP 和 X 射线图像组合、（4）CT-AP 和 X 射线图像下采样组合以及（5）CT-AP 图像对预训练的 ResNet18 进行了 rHOA 与非 rHOA 的分类任务优化。
在五个模型中，模型 3 和模型 5 在从 CT-AP 图像中检测 rHOA 方面表现最佳。模型 3 在 CT-AP 图像测试集上检测 rHOA 的平衡准确率为 82.2%，能够以 0.93（0.75-0.99）的受试者工作特征曲线下面积（ROC AUC）区分 rHOA 与非 rHOA。模型 5 在 CT-AP 图像测试集上检测 rHOA 的准确率为 82.2%，以 0.89（0.67-0.97）的 ROC AUC 区分 rHOA 与非 rHOA。
类似于 X 射线的 CT 合成图像可用于检测 rHOA。此外，在缺乏大量训练数据的情况下，可以通过结合 CT-AP 和 X 射线图像来优化可靠的 DL 模型。
