Fully automatic knee osteoarthritis severity grading using deep neural networks with a novel ordinal loss.
Fully automatic knee osteoarthritis severity grading using deep neural networks with a novel ordinal loss.
复制标题
DOI：
10.1016/j.compmedimag.2019.06.002
复制
发表时间：
2019-07
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
膝骨关节炎（OA）是老年人活动受限和身体残疾的主要原因之一。早期发现和干预有助于减缓骨关节炎的退化。医生基于目视检查的评分是主观的，不同译员的评分也不同，并且高度依赖于他们的经验。在本文中，我们连续应用两个深度卷积神经网络（CNN）来自动测量膝关节 OA 的严重程度，并通过 Kellgren-Lawrence（KL）分级系统进行评估。首先，考虑到 X 射线图像中膝关节大小分布的变异性较小，我们使用定制的一阶段 YOLOv2 网络检测膝关节。其次，我们对最流行的 CNN 模型（包括 ResNet、VGG、DenseNet 以及 InceptionV3 的变体）进行微调，以使用新颖的可调整序数损失对检测到的膝关节图像进行分类。具体来说，受膝盖 KL 评分任务的序数性质的启发，我们对预测 KL 评分与真实 KL 评分之间距离较大的错误分类分配更高的惩罚。来自骨关节炎倡议 (OAI) 数据集的基线 X 射线图像用于评估。在膝关节检测上，我们在 Jaccard 指数阈值 0.75 下实现了平均 Jaccard 指数 0.858 和 92.2% 的召回率。在knee KL分级任务上，采用所提出的序数损失的微调VGG-19模型获得了69.7%的最佳分类精度和0.344的平均绝对误差（MAE）。膝关节检测和膝关节 KL 分级均实现了最先进的性能。代码、数据集和模型发布于 https://github.com/PingjunChen/KneeAnalysis。
Knee osteoarthritis (OA) is one major cause of activity limitation and physical disability in older adults. Early detection and intervention can help slow down the OA degeneration. Physicians’ grading based on visual inspection is subjective, varied across interpreters, and highly relied on their experience. In this paper, we successively apply two deep convolutional neural networks (CNN) to automatically measure the knee OA severity, as assessed by the Kellgren-Lawrence (KL) grading system. Firstly, considering the size of knee joints distributed in X-ray images with small variability, we detect knee joints using a customized one-stage YOLOv2 network. Secondly, we fine-tune the most popular CNN models, including variants of ResNet, VGG, and DenseNet as well as InceptionV3, to classify the detected knee joint images with a novel adjustable ordinal loss. To be specific, motivated by the ordinal nature of the knee KL grading task, we assign higher penalty to misclassification with larger distance between the predicted KL grade and the real KL grade. The baseline X-ray images from the Osteoarthritis Initiative (OAI) dataset are used for evaluation. On the knee joint detection, we achieve mean Jaccard index of 0.858 and recall of 92.2% under the Jaccard index threshold of 0.75. On the knee KL grading task, the fine-tuned VGG-19 model with the proposed ordinal loss obtains the best classification accuracy of 69.7% and mean absolute error (MAE) of 0.344. Both knee joint detection and knee KL grading achieve state-of-the-art performance. The code, dataset, and models are released at https://github.com/PingjunChen/KneeAnalysis.
登录
查看更多内容
DOI：
10.1016/j.compmedimag.2018.04.005
发表时间：
2018-09-01
影响因子：
5.7
作者：
Jiang, Zhexin;Zhang, Hao;Ko, Seok-Bum
通讯作者：
Ko, Seok-Bum
影响因子：
3.4
作者：
Conaghan, Philip G.;Porcheret, Mark;Birrell, Fraser
通讯作者：
Birrell, Fraser
影响因子：
7
作者：
Neogi, T.
通讯作者：
Neogi, T.
影响因子：
64.8
作者：
Esteva A;Kuprel B;Novoa RA;Ko J;Swetter SM;Blau HM;Thrun S
通讯作者：
Thrun S
DOI：
10.1016/j.compmedimag.2018.10.007
发表时间：
2019-01-01
影响因子：
5.7
作者：
Mahbod, Amirreza;Schaefer, Gerald;Wang, Chunliang
通讯作者：
Wang, Chunliang