Automatic segmentation model of intercondylar fossa based on deep learning: a novel and effective assessment method for the notch volume.

BACKGROUND: Notch volume is associated with anterior cruciate ligament (ACL) injury. Manual tracking of intercondylar notch on MR images is time-consuming and laborious. Deep learning has become a powerful tool for processing medical images. This study aims to develop an MRI segmentation model of intercondylar fossa based on deep learning to automatically measure notch volume, and explore its correlation with ACL injury. METHODS: The MRI data of 363 subjects (311 males and 52 females) with ACL injuries incurred during non-contact sports and 232 subjects (147 males and 85 females) with intact ACL were retrospectively analyzed. Each layer of intercondylar fossa was manually traced by radiologists on axial MR images. Notch volume was then calculated. We constructed an automatic segmentation system based on the architecture of Res-UNet for intercondylar fossa and used dice similarity coefficient (DSC) to compare the performance of segmentation systems by different networks. Unpaired t-test was performed to determine differences in notch volume between ACL-injured and intact groups, and between males and females. RESULTS: The DSCs of intercondylar fossa based on different networks were all more than 0.90, and Res-UNet showed the best performance. The notch volume was significantly lower in the ACL-injured group than in the control group (6.12 ± 1.34 cm3 vs. 6.95 ± 1.75 cm3, P < 0.001). Females had lower notch volume than males (5.41 ± 1.30 cm3 vs. 6.76 ± 1.51 cm3, P < 0.001). Males and females who had ACL injuries had smaller notch than those with intact ACL (p < 0.001 and p < 0.005). Men had larger notches than women, regardless of the ACL injuries (p < 0.001). CONCLUSION: Using a deep neural network to segment intercondylar fossa automatically provides a technical support for the clinical prediction and prevention of ACL injury and re-injury after surgery.

Improving sensitivity and connectivity of retinal vessel segmentation via error discrimination network.

PURPOSE: Automated retinal vessel segmentation is crucial to the early diagnosis and treatment of ophthalmological diseases. Many deep-learning-based methods have shown exceptional success in this task. However, current approaches are still inadequate in challenging vessels (e.g., thin vessels) and rarely focus on the connectivity of vessel segmentation. METHODS: We propose using an error discrimination network (D) to distinguish whether the vessel pixel predictions of the segmentation network (S) are correct, and S is trained to obtain fewer error predictions of D. Our method is similar to, but not the same as, the generative adversarial network. Three types of vessel samples and corresponding error masks are used to train D, as follows: (1) vessel ground truth; (2) vessel segmented by S; (3) artificial thin vessel error samples that further improve the sensitivity of D to wrong small vessels. As an auxiliary loss function of S, D strengthens the supervision of difficult vessels. Optionally, we can use the errors predicted by D to correct the segmentation result of S. RESULTS: Compared with state-of-the-art methods, our method achieves the highest scores in sensitivity (86.19%, 86.26%, and 86.53%) and G-Mean (91.94%, 91.30%, and 92.76%) on three public datasets, namely, STARE, DRIVE, and HRF. Our method also maintains a competitive level in other metrics. On the STARE dataset, the F1-score and area under the receiver operating characteristic curve (AUC) of our method rank second and first, respectively, reaching 84.51% and 98.97%. The top scores of the three topology-relevant metrics (Conn, Inf, and Cor) demonstrate that the vessels extracted by our method have excellent connectivity. We also validate the effectiveness of error discrimination supervision and artificial error sample training through ablation experiments. CONCLUSIONS: The proposed method provides an accurate and robust solution for difficult vessel segmentation.

Identity-mapping cascaded network for fMRI registration.

Neuroscience researches based on functional magnetic resonance imaging (fMRI) rely on accurate inter-subject image registration of functional regions. The intersubject alignment of fMRI can improve the statistical power of group analyses. Recent studies have shown the deep learning-based registration methods can be used for registration. In our work, we proposed a 30-Identity-Mapping Cascaded network (30-IMCNet) for rs-fMRI registration. It is a cascaded network that can warp the moving image progressively and finally align to the fixed image. A Combination unit with an identity-mapping path is added to the inputs of each IMCNet to guide the network training. We implemented 30-IMCNet on an rs-fMRI dataset (1000 Functional Connectomes Project dataset) and a task-related fMRI dataset (Eyes Open Eyes Closed fMRI dataset). To evaluate our method, a group-level analysis was implemented in the testing dataset. For rs-fMRI, the criterions such as peakt-value of group-level t-maps, cluster-level evaluation, and intersubject functional network correlation were used to evaluate the quality of the registrations. For task-related fMRI, peakt-value in ALFF paired-t map and peakt-value in ReHo paired-t maps were used. Compared with traditional algorithm FSL, SPM, and deep learning algorithm Kimet al, Zhaoet alour method has improvements of 48.90%, 30.73%, 36.38%, and 16.73% in the peaktvalue of t-maps. Our proposed method can achieve superior functional registration performance and thus gain a significant improvement in functional consistency.