Abstract:CD317 (Tetherin, BST-2 or HM1.24), encoded by the BST-2 gene, is a type II transmembrane glycoprotein with a unique topology. CD317 is constitutively expressed in a variety of human tissues, and can also be induced by some cytokines such as interferon. Recently, many studies have shown that CD317 is overexpressed in different types of tumors with critical roles to facilitate tumor progression by regulating multiple biological processes, such as proliferation, migration and apoptosis. Therefore, CD317 is suggested as a promising target for tumor therapy. This review focuses on the research progress of CD317 in tumor development, including its expression within tumors, mechanisms for tumorigenesis, and potential as a target against cancer, and will hopefully provide novel ideas and direction for discovering tumorigenesis is overexpressed in different types of tumors with critical roles to facilitate tumor progression by regulating multiple biological processes, such as proliferation, migration and apoptosis. Therefore, CD317 is suggested as a promising target for tumor therapy. This review focuses on the research progress of CD317 in tumor development, including its expression within tumors, mechanisms for tumorigenesis, and potential as a target against cancer, and will hopefully provide novel ideas and direction for discovering tumorigenesis mechanisms and new therapeutic strategies.

Abstract:A novel cell analyzer and a component biochip, based on microfluidic technology and Coulter principle, was developed to count leukocyte cells for point-of-care testing. The microfluidic chip was composed of multi-layer structure, and the liquid flowed between the upper and lower layers of the chip, not in contact with the cell analyzer. The cell analyzer and chips were used to test standard particles and leukocyte cells to characterize the performance. The results showed that the variation coefficients of the standard particle counting results are 4.95%, 3.56% and 2.13% with different diameters, respectively. The total leukocyte count results by the microfluidic cell analyzer and hematology analyzer Sysmex XN20 were compared. The linear regression coefficient R2 is 0.999 7. Bland-Altman plots showed that more than 95% of points fluctuate within the 95% consistency bound. The expected bias percentages at the three medical decision levels of 3.00×109/L, 11.00×109/L and 30.00×109/L were 2.12%, 0.54%, and 0.17%, respectively. Thus, the cell analyzer has good repeatability for standard particle test, and the results of leukocyte test are comparable with those of XN20. It can be used to detect leukocyte in clinical blood sample, especially suitable for village clinics and other basic medical units.

Abstract:In order to solve rough contour of some blood vessels and the loss of vessel-perpherals and branches in traditional retinal vessel segmentation, a novel method forretinal vessel segmentation which combines linear spectral clustering super-pixel with generative adversarial networks (GAN) is proposed.The accuracy of segmentation is improved using the multi-scale features from atrousspatial pyramid pooling (ASPP) module with a modified GAN method. After the segmentation image is obtained, by utilizing the characteristics of high edge suitability and clear contour of linear spectral clustering super-pixel segmentation, the GAN output image was mapped to the super-pixel segmentation image. The segmentation was achieved by classifying the pixel clusters. The experimental results show that compared with the traditional retinal vessel segmentation method, the sensitivity and accuracy of the proposed method are improved, especially in the details of the contour edge.

Abstract:Medical image-guided percutaneous renal puncture is an important technique to establish surgical channels in the percutaneous nephrolithotomy. High quality and real-time medical image can improve the accuracy of intraoperative puncture and reduce surgical risks. Aiming at the navigation and positioning problem of percutaneous renal puncture under free breathing, a renal puncture positioning method based on the intraoperative 2D ultrasound and the preoperative 3D CT image is introduced. In the first step, a real-time registration method was applied to the 2D ultrasound and the preoperative 3D CT image to realize rapid registration and fusion. Then the fusion image is registered in the surgical space through ultrasound probe calibration and spatial coordinate transformation, which realized the real-time positioning of puncture targets. Finally, the human abdomen model was used for the experiments, and the results showed that the root mean square error of the ultrasound probe calibration is 0.998 mm. The registration error of 2D ultrasound and 3D CT images is 0.709 mm, the average registration time was 1.15 s, and the final average positioning error is 2.265 mm.

Abstract:Accurate segmentation of COVID-19 pneumonia lesions on chest CT images can facilitate the diagnosis of pneumonia. The CT image finds of which contained the ground-glass opacity, consolidation, pleural effusion, etc. This study proposed a deep neural network RCB-UNet＋＋ for the segmentation of COVID-19 pneumonia lesions in CT images, which exhibit large variations in texture, size and location. The model was built on top of the UNet＋＋ network with an extra residual module and an attention module. This architecture is able to effectively extract low-level texture features and high-level semantic information, thus improving the model performance. The RCB-UNet＋＋ model was trained on 45 samples and tested by another 50 cases. Finally, it achieved a Dice coefficient of 0.715, a sensitivity and specificity of 0.754 and 0.952, outperforming other designed models on the same dataset. The results demonstrate that the proposed algorithm improves the segmentation performance and has potential in fully automatic segmentation of COVID-19 pneumonia lesions on CT images.

Abstract:We propose a novel learning strategy inspired by domain decomposition methods to accelerate the training of convolutional neural network (CNN). The proposed method is applied to residual networks (ResNet) for image classification tasks. The best result is achieved with ResNet32. In this case, we split ResNet32 into 4 sub-networks. Each sub-network has 0.47 M parameters which is 1/16 of the original ResNet32, thereby facilitating the learning process. Moreover, because the sub-networks can be trained in parallel, the computational time can therefore be reduced to 5.65 h from 8.53 h (by the conventional learning strategy) in the classification task with the CIFAR-10 dataset. We also find that the accuracy of the classification is improved to 94.09% from 92.82%. Similar improvements are also achieved with the CIFAR-100 and Food-101 datasets. In conclusion, the proposed learning strategy can reduce the computational time substantially with improved accuracy in classification. The results suggest that the proposed strategy can potentially be applied to train CNN with a large amount of parameters.

Abstract:To solve the problem of serious imbalance between the foreground and background in medical images and small objects segmentation, we propose an attention network based on Gaussian image pyramid to fuse spatial information and abstract information in the feature decoding stage. In addition, a feature recaller is designed to force the encoder to avoid missing features of the region of interest. Finally, a hybrid loss function composed of classification accuracy and global overlapping terms is employed to deal with the serious imbalance between the foreground and background. The proposed method was validated on a knee articular cartilage dataset and the COVOID-19 chest CT dataset where the foreground proportions are 2.08% and 10.73%, respectively. The proposed method achieves the highest Dice coefficients on both datasets as compared with U-Net and its state-of-the-art variants, which are 0.884±0.032 and 0.831±0.072, respectively.

Abstract:Thanks to the rapid evolvement of technologies including big data and cloud computing, application systems become more and more centralized with boosted scale, which gradually highlights the performance issues of storage systems. Parallel file systems have been applied in a wide range of applications to meet the performance requirements of large-scale applications running on the storage systems. However, the majority of currently used parallel file system optimization methods only takes the application system or the parallel file system itself into account, and seldom considers the collaboration among them. Considering that the access mode of an application system when accessing the parallel file system will have a significant impact on the storage system performance, this study proposes a parallel file system optimization approach based on dynamic partitioning. The key idea is to firstly leverage machine learning techniques to reveal the relationships between factors that can influence the system performance and build an optimization model accordingly. Then, the optimization model will facilitate the parameter optimization of parallel file systems. Finally, the model is tested on a Ceph-based storage system prototype with a three-layer application system. The proposed model successfully optimizes the parallel file system access performance. Experimentally, the proposed model achieves an optimization prediction accuracy of 85%. With the assistance of the proposed model, the system throughput is improved by 3.6 times.

Abstract:

Abstract:

Abstract: