Deep learning-based fully automatic segmentation of wrist cartilage in MR images.
Deep learning-based fully automatic segmentation of wrist cartilage in MR images.
复制标题
DOI：
10.1002/nbm.4320
复制
发表时间：
2020-08
影响因子：
2.9
通讯作者：
Andreychenko A
中科院分区：
文献类型：
--
作者：
Brui E;Efimtcev AY;Fokin VA;Fernandez R;Levchuk AG;Ogier AC;Samsonov AA;Mattei JP;Melchakova IV;Bendahan D;Andreychenko A
该研究的目的是研究专用卷积神经网络（CNN）的性能，该网络针对2D MR图像的腕关节软骨分割进行了优化。CNN利用平面架构和基于补丁（PB）的训练方法，确保在存在有限数量的训练数据的情况下获得最佳性能。CNN在11名受试者（健康志愿者和患者）中使用两种不同线圈采集的20个多层MRI数据集中进行了训练和验证。验证包括与用于从MR图像分割关节的替代最先进CNN方法和地面实况手动分割的比较。当在有限的训练数据上训练时，CNN的表现明显优于基于图像和基于补丁的U-Net网络。我们的PB-CNN还在具有大量软骨组织的代表性（中央冠状）切片中表现出与手动分割（Sørensen-Dice相似系数（DSC）= 0.81）的良好一致性。对于软骨组织数量非常有限的切片，网络性能降低，这表明需要完全3D卷积网络来提供跨关节的统一性能。该研究还评估了手动腕关节软骨分割的观察者间和观察者内变异性（DSC分别为0.78-0.88和0.9）。提出的基于深度学习的MRI腕关节软骨分割可以促进腕关节骨关节炎新成像标志物的研究，以表征其进展和对治疗的反应。
The study objective was to investigate the performance of a dedicated convolutional neural network (CNN) optimized for wrist cartilage segmentation from 2D MR images. CNN utilized a planar architecture and patch-based (PB) training approach that ensured optimal performance in the presence of a limited amount of training data. The CNN was trained and validated in twenty multi-slice MRI datasets acquired with two different coils in eleven subjects (healthy volunteers and patients). The validation included a comparison with the alternative state-of-the-art CNN methods for the segmentation of joints from MR images and the ground-truth manual segmentation. When trained on the limited training data, the CNN outperformed significantly image-based and patch-based U-Net networks. Our PB-CNN also demonstrated a good agreement with manual segmentation (Sørensen–Dice similarity coefficient (DSC) = 0.81) in the representative (central coronal) slices with large amount of cartilage tissue. Reduced performance of the network for slices with a very limited amount of cartilage tissue suggests the need for fully 3D convolutional networks to provide uniform performance across the joint. The study also assessed inter- and intra-observer variability of the manual wrist cartilage segmentation (DSC=0.78–0.88 and 0.9, respectively). The proposed deep-learning-based segmentation of the wrist cartilage from MRI could facilitate research of novel imaging markers of wrist osteoarthritis to characterize its progression and response to therapy.
登录
查看更多内容
影响因子：
1.7
作者：
Nishimura, K;Tanabe, T;Matsushita, T
通讯作者：
Matsushita, T
影响因子：
19.7
作者：
PETERFY, CG;VANDIJKE, CF;GENANT, HK
通讯作者：
GENANT, HK
影响因子：
4.8
作者：
DICE, LR
通讯作者：
DICE, LR
DOI：
10.1007/978-3-319-46723-8_40
发表时间：
2016-10-01
期刊：
Medical image computing and computer-assisted intervention : MICCAI ... International Conference on Medical Image Computing and Computer-Assisted Intervention
影响因子：
--
作者：
Kashyap, Satyananda;Oguz, Ipek;Sonka, Milan
通讯作者：
Sonka, Milan
影响因子：
0.7
作者：
Li, Angela E.;Lee, Steve K.;Wolfe, Scott W.
通讯作者：
Wolfe, Scott W.