The objective of this study is to develop, test and validate a fully automatic, deep learning-based segmentation method for the wrist joint cartilage in magnetic resonance images. The study was conducted in 8 healthy volunteers and 3 patients with wrist joint diseases. 3D MRI datasets (20 in total) were acquired at 1.5T using a VIBE sequence. Wrist cartilage was segmented on coronal slices by a clinician and the convolutional neural network (CNN) was trained, developed and tested using the corresponding segmented masks. For an inter and intra observer study wrist cartilage was segmented by three observers once and twice by one observer on a dataset of 20 central coronal slices. Performance of the CNN was compared quantitatively to the manual segmentations using the concordance and the Sørensen-Dice similarity coefficients (DSC). Cartilage segmentations obtained with the CNN showed a substantial agreement with the manual segmentations for the whole wrist joint (DSC = 0.73) and a good agreement (DSC = 0.81) for the central coronal slices. The inter-and intra-observer concordance indices for manual segmentations were 0.55 and 0.85, respectively. The concordance index of the CNN-based segmentation was 0.69 when compared to the manual segmentations. The fully automatic deep-learning based segmentation of the wrist cartilage showed a high concordance with the manual measurements. It could be applied to determine an automatic, quantitative metric in clinical wrist cartilage studies.
Purpose: Automatic measurement of wrist cartilage volume in MR images.
Methods:We assessed the performance of four manually optimized variants of the U-Net architecture, nnU-Net and Mask R-CNN frameworks for the segmentation of wrist cartilage. The results were compared to those from a patch-based convolutional neural network (CNN) we previously designed. The segmentation quality was assessed on the basis of a comparative analysis with manual segmentation. The best networks were compared using a cross-validation approach on a dataset of 33 3D VIBE images of mostly healthy volunteers. Influence of some image parameters on the segmentation reproducibility was assessed. Results: The U-Net-based networks outperformed the patch-based CNN in terms of segmentation homogeneity and quality, while Mask R-CNN did not show an acceptable performance. The median 3D DSC value computed with the U-Net_AL (0.817) was significantly larger than DSC values computed with the other networks. In addition, the U-Net_AL provided the lowest mean volume error (17%) and the highest Pearson correlation coefficient (0.765) with respect to the ground truth values. Of interest, the reproducibility computed using U-Net_AL was larger than the reproducibility of the manual segmentation. Moreover, the results indicate that the MRI-based wrist cartilage volume is strongly affected by the image resolution.Conclusions: U-Net CNN with attention layers provided the best wrist cartilage segmentation performance. In order to be used in clinical conditions, the trained network can be fine-tuned on a dataset representing a group of specific patients.The error of cartilage volume measurement should be assessed independently using a non-MRI method.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.