Search

The combination of a blood cell analyzer and artificial microscopy to detect white blood cells is used in hospitals. Blood cell analyzers not only have large throughput, but they also cannot detect cell morphology; although artificial microscopy has high accuracy, it is inefficient and prone to missed detections. In view of the above problems, a method based on Fourier ptychographic microscopy (FPM) and deep learning to detect peripheral blood leukocytes is proposed in this paper. Firstly, high-resolution and wide-field microscopic images of human peripheral blood cells are obtained using the FPM system, and the cell image data are enhanced with DCGANs (deep convolution generative adversarial networks) to construct datasets for performance evaluation. Then, an improved DETR (detection transformer) algorithm is proposed to improve the detection accuracy of small white blood cell targets; that is, the residual module Conv Block in the feature extraction part of the DETR network is improved to reduce the problem of information loss caused by downsampling. Finally, CIOU (complete intersection over union) is introduced as the bounding box loss function, which avoids the problem that GIOU (generalized intersection over union) is difficult to optimize when the two boxes are far away and the convergence speed is faster. The experimental results show that the mAP of the improved DETR algorithm in the detection of human peripheral white blood cells is 0.936. In addition, this algorithm is compared with other convolutional neural networks in terms of average accuracy, parameters, and number of inference frames per second, which verifies the feasibility of this method in microscopic medical image detection.

Seeking good correspondences between two images is a fundamental and challenging problem in the remote sensing (RS) community, and it is a critical prerequisite in a wide range of feature-based visual tasks. In this article, we propose a flexible and general deep state learning network for both rigid and nonrigid feature matching, which provides a mechanism to change the state of matches into latent canonical forms, thereby weakening the degree of randomness in matching patterns. Different from the current conventional strategies (i.e., imposing a global geometric constraint or designing additional handcrafted descriptor), the proposed StateNet is designed to perform alternating two steps: 1) recalibrates matchwise feature responses in the spatial domain and 2) leverages the spatially local correlation across two sets of feature points for transformation update. For this purpose, our network contains two novel operations: adaptive dual-aggregation convolution (ADAConv) and point rendering layer (PRL). These two operations are differentiable, so our network can be inserted into the existing classification architecture to reduce the cost of establishing reliable correspondences. To demonstrate the robustness and universality of our approach, extensive experiments on various real image pairs for feature matching are conducted. Experiments reveal the superiority of our StateNet significantly over the state-of-the-art alternatives.

The transmission of satellite payload data is critical for services provided by aerospace ground networks. To ensure the correctness of data transmission, the TCP data transmission protocol has been used typically. However, the standard TCP congestion control algorithm is incompatible with networks with a long time delay and a large bandwidth, resulting in low throughput and resource waste. This article compares recent studies on TCP-based acceleration algorithms and proposes an acceleration algorithm based on the learning of historical characteristics, such as end-to-end delay and its variation characteristics, the arrival interval of feedback packets (ACK) at the receiving end and its variation characteristics, the degree of data packet reversal and its variation characteristics, delay and jitter caused by the security equipment's deep data inspection, and random packet loss caused by various factors. The proposed algorithm is evaluated and compared with the TCP congestion control algorithms under both laboratory and ground network conditions. Experimental results indicate that the proposed acceleration algorithm is efficient and can significantly increase throughput. Therefore, it has a promising application prospect in high-speed data transmission in aerospace-ground service networks.

Protein-ligand blind docking is a powerful method for exploring the binding sites of receptors and the corresponding binding poses of ligands. It has seen wide applications in pharmaceutical and biological researches. Previously, we proposed a blind docking server, CB-Dock, which has been under heavy use (over 200 submissions per day) by researchers worldwide since 2019. Here, we substantially improved the docking method by combining CB-Dock with our template-based docking engine to enhance the accuracy in binding site identification and binding pose prediction. In the benchmark tests, it yielded the success rate of ∼85% for binding pose prediction (RMSD < 2.0 Å), which outperformed original CB-Dock and most popular blind docking tools. This updated docking server, named CB-Dock2, reconfigured the input and output web interfaces, together with a highly automatic docking pipeline, making it a particularly efficient and easy-to-use tool for the bioinformatics and cheminformatics communities. The web server is freely available at https://cadd.labshare.cn/cb-dock2/., (© The Author(s) 2022. Published by Oxford University Press on behalf of Nucleic Acids Research.)

In the domain of point cloud registration, the coarse-to-fine feature matching paradigm has received significant attention due to its impressive performance. This paradigm involves a two-step process: first, the extraction of multilevel features, and subsequently, the propagation of correspondences from coarse to fine levels. However, this approach faces two notable limitations. Firstly, the use of the Dual Softmax operation may promote one-to-one correspondences between superpoints, inadvertently excluding valuable correspondences. Secondly, it is crucial to closely examine the overlapping areas between point clouds, as only correspondences within these regions decisively determine the actual transformation. Considering these issues, we propose OAAFormer to enhance correspondence quality. On the one hand, we introduce a soft matching mechanism to facilitate the propagation of potentially valuable correspondences from coarse to fine levels. On the other hand, we integrate an overlapping region detection module to minimize mismatches to the greatest extent possible. Furthermore, we introduce a region-wise attention module with linear complexity during the fine-level matching phase, designed to enhance the discriminative capabilities of the extracted features. Tests on the challenging 3DLoMatch benchmark demonstrate that our approach leads to a substantial increase of about 7% in the inlier ratio, as well as an enhancement of 2%–4% in registration recall. Finally, to accelerate the prediction process, we replace the Conventional Random Sample Consensus (RANSAC) algorithm with the selection of a limited yet representative set of high-confidence correspondences, resulting in a 100 times speedup while still maintaining comparable registration performance. [ABSTRACT FROM AUTHOR]

Targeting the issues of slow speed and inadequate precision of optimal solution calculation for multi-user detection in complex noise environments, this paper proposes a multi-user detection algorithm based on a Hybrid Cheetah Optimizer (HCO). The algorithm first optimizes the control parameters and individual update mechanism of the Cheetah Optimizer (CO) algorithm using a nonlinear strategy to improve the uniformity and discretization of the individual search range, and then dynamically introduces a differential evolutionary algorithm into the improved selection mechanism of the CO algorithm, which is utilized to fine-tune the solution space and maintain the local diversity during the fast search process. Simulation results demonstrate that this detection algorithm not only realizes fast convergence with a very low bit error rate (BER) at eight iterations but also has obvious advantages in terms of noise immunity, resistance to far and near effects, communication capacity, etc., which greatly improves the speed and accuracy of optimal position solving for multi-user detection and can achieve the purpose of accurate solving in complex environments. [ABSTRACT FROM AUTHOR]

Breast cancer (BC) is the most commonly diagnosed non‐skin cancer in women. To achieve early and accurate diagnosis, three‐dimensional (3D) cone‐beam breast computed tomography (CBBCT) and digital breast tomosynthesis (DBT) modalities are used. Importantly, the comparison of reconstruction accuracy of both CBBCT and DBT has rarely been investigated, thus constituting a research gap. This study systematically investigated the performances of the CBBCT and DBT for different reconstruction algorithms using both BR3D breast phantom and breast images. We acquired clinical breast images and scanned the BR3D phantom for additional breast images. These acquired images were used to simulate projection data using predefined CBBCT and DBT geometries. The simulated projections were reconstructed using five different reconstruction algorithms. To evaluate the reconstruction accuracy, we calculated average image quality assessment (IQA) indices, including peak signal‐to‐noise ratio (PSNR), structural similarity index (SSIM), root mean square error (RMSE), and others, across different algorithms and modalities. The pooled PSNR, SSIM, and RMSE for DBT and CBBCT images are (31.6265 ± 0.8725), (0.9353 ± 0.0077), and (0.0270 ± 0.0025) and (29.7007 ± 0.9249), (0.9136 ± 0.0130), and (0.0342 ± 0.0040), which implies that the overall IQA indices of DBT are superior to CBBCT; therefore, DBT tends to reveal more BC detectability as the diagnosis outcome would largely depend on good quality images. The results show that DBT gives an improved result for all algorithms compared to CBBCT, although further experimental trials may be needed to establish the findings fully. The findings suggest that using DBT may enhance the accuracy of BC diagnosis compared to CBBCT due to its superior image quality in clinical practice, emphasizing the importance of selecting optimal reconstruction algorithms for improved diagnostic outcomes. [ABSTRACT FROM AUTHOR]

Due to the limited memory and computing resources in the real application of target detection, the method is challenging to implement on mobile and embedded devices. In order to achieve the balance between detection accuracy and speed in pedestrian-intensive scenes, an improved lightweight dense pedestrian detection algorithm GS-YOLOv5 (GhostNet GSConv- SIoU) is proposed in this paper. In the Backbone section, GhostNet is used to replace the original CSPDarknet53 network structure, reducing the number of parameters and computation. The CBL module is replaced with GSConv in the Head section, and the CSP module is replaced with VoV-GSCSP. The SloU loss function is used to replace the original IoU loss function to improve the prediction box overlap problem in dense scenes. The model parameters are reduced by 40% and the calculation amount is reduced by 64% without losing the average accuracy, and the detection accuracy is improved by 0.5%. The experimental results show that the GS-YOLOv5 can detect pedestrians more effectively under limited hardware conditions to cope with dense pedestrian scenes, and it is suitable for the online real-time detection of pedestrians. [ABSTRACT FROM AUTHOR]

In the process of two‐phase countercurrent flow in the extraction column, the study of the hydrodynamic characteristics of dispersed phase droplets is crucial for the design and scaling up of the extraction column. In a pulsed extraction column with a liquid–liquid system, the four‐sensor optical fiber probe was used to assess the hydrodynamic characteristics of the dispersed phase droplets, including size, mean velocity, and holdup. 1 M HNO3 and 30% TBP/kerosene were used as continuous phase and dispersed phase, respectively. The effects of two‐phase superficial velocities and pulsation intensity were investigated on droplet size, mean velocity, and holdup. The findings show that aside from the local holdup, the pulsation intensity has the greatest impact on the hydrodynamic characteristics. A new and concise algorithm was used to deal with the light signals to determine the droplet size and mean velocity. This algorithm created two Cartesian coordinate systems, each with a separate z‐axis using the leading probe as the origin of the coordinate instead of using the vector approach. Compared to the previous algorithm, both the needed number of the time interval between the voltage peaks and that of the direction angles decrease. The relative error of the droplet size obtained by this improved algorithm is within 7%, and droplet velocity is almost the same, compared to the previous algorithm, which indicates that the new algorithm of the four‐sensor optical fiber signal processing is reliable to the measurement of the hydrodynamic characteristics. [ABSTRACT FROM AUTHOR]

A class of proximal gradient-type algorithm for bilevel nonlinear nondifferentiable programming problems with smooth substructure is developed in this paper. The original problem is approximately reformulated by explicit slow control technique to a parameterized family function which makes full use of the information of smoothness. At each iteration, we only need to calculate one proximal point analytically or with low computational cost. We prove that the accumulation iterations generated by the algorithms are solutions of the original problem. Moreover, some results of complexity of the algorithms are presented in convergence analysis. Numerical experiments are implemented to verify the efficiency of the proximal gradient algorithms for solving this kind of bilevel programming problems. [ABSTRACT FROM AUTHOR]

publisher‐imprint‐name Springer volume‐issue‐count 1 issue‐article‐count 0 issue‐toc‐levels 0 issue‐pricelist‐year 2020 issue‐copyright‐holder The Author(s) issue‐copyright‐year 2020 article‐contains‐esm Yes article‐numbering‐style Unnumbered article‐registration‐date‐year 2020 article‐registration‐date‐month 1 article‐registration‐date‐day 20 article‐toc‐levels 0 toc‐levels 0 volume‐type Regular journal‐product ArchiveJournal numbering‐style Unnumbered article‐grants‐type OpenChoice metadata‐grant OpenAccess abstract‐grant OpenAccess bodypdf‐grant OpenAccess bodyhtml‐grant OpenAccess bibliography‐grant OpenAccess esm‐grant OpenAccess online‐first false pdf‐file‐reference BodyRef/PDF/40169_2020_Article_263.pdf pdf‐type Typeset target‐type OnlinePDF issue‐type Regular article‐type OriginalPaper journal‐subject‐primary Medicine & Public Health journal‐subject‐secondary Medicine/Public Health, general journal‐subject‐collection Medicine open‐access true --> Background and Aim: To develop and validate radiomic prediction models using contrast‐enhanced computed tomography (CE‐CT) to preoperatively predict Ki‐67 expression in gastrointestinal stromal tumors (GISTs). Method: A total of 339 GIST patients from four centers were categorized into the training, internal validation, and external validation cohort. By filtering unstable features, minimum redundancy, maximum relevance, Least Absolute Shrinkage and Selection Operator (LASSO) algorithm, a radiomic signature was built to predict the malignant potential of GISTs. Individual nomograms of Ki‐67 expression incorporating the radiomic signature or clinical factors were developed using the multivariate logistic model and evaluated regarding its calibration, discrimination, and clinical usefulness. Results: The radiomic signature, consisting of 6 radiomic features had AUC of 0.787 [95% confidence interval (CI) 0.632–0.801], 0.765 (95% CI 0.683–0.847), and 0.754 (95% CI 0.666–0.842) in the prediction of high Ki‐67 expression in the training, internal validation and external validation cohort, respectively. The radiomic nomogram including the radiomic signature and tumor size demonstrated significant calibration, and discrimination with AUC of 0.801 (95% CI 0.726–0.876), 0.828 (95% CI 0.681–0.974), and 0.784 (95% CI 0.701–0.868) in the training, internal validation and external validation cohort respectively. Based on the Decision curve analysis, the radiomics nomogram was found to be clinically significant and useful. Conclusions: The radiomic signature from CE‐CT was significantly associated with Ki‐67 expression in GISTs. A nomogram consisted of radiomic signature, and tumor size had maximum accuracy in the prediction of Ki‐67 expression in GISTs. Results from our study provide vital insight to make important preoperative clinical decisions. [ABSTRACT FROM AUTHOR]

With the increasing popularity of Internet of things (IoT) applications, mobile edge computing (MEC) is emerging as a new technology. Caching popular content to edge servers can effectively reduce backhaul time and satisfy a large number of users' access requests for the content. However, edge caching faces the problems of limited storage capacity in the edge server and limited service scope in MEC. Therefore, edge caching should consider allocating the limited caching resources reasonably to different content providers (CP) to obtain the high caching utility. In this paper, we study caching in a multi-CP scenario with multiple users. In the process of defining content popularity, we consider that CPs′ overall popularity degree satisfies the law of diminishing marginal effect, and define the user satisfaction function as the benefit of caching. Then, aiming at maximizing caching utility, we establish a game model of caching resource competition among CPs based on noncooperative game, and prove the existence of Nash equilibrium (NE). In addition, the Best Response based Multi-CP Caching (BRMC) algorithm is proposed to obtain the best caching strategy. Finally, iterative experiments validate the convergence of the BRMC, and comparative experiments show that the BRMC can achieve a high caching utility. [ABSTRACT FROM AUTHOR]

Owing to the comprehensive effects of dimensional disaster, interpolation error, and Markov characteristics of the controlled objects, the traditional dynamic programming algorithm has difficulty ensuring the efficiency of calculation, the accuracy of the results, and the rationality of the optimal control law when solving the optimal fuel economy problem for a multi-mode and multi-gear hybrid electric vehicles (HEVs). To solve this problem, an improved dynamic programming algorithm, CQU-DP, is proposed herein. The algorithm can rapidly generate optimal solutions with high accuracy and rationality through grid size configuration, matrix expansion, filtering, and the introduction of state and control variable penalties. The optimal fuel economy of a parallel HEV with five gears and six modes was solved using this algorithm. The results indicated that compared with the traditional basic dynamic programming algorithm (B-DP) and an improved dynamic programming algorithm (SJTU-DP), the proposed optimization algorithm reduced the calculation time by 96.36% and 93.79%, and the fuel economy was increased by 26.63% and 1.92%, respectively. Additionally, the optimal control law was more reasonable. • Global optimal economy solution of hybrid electric vehicles (HEVs). • Account efficiency, accuracy, and rationality. • A fast dynamic-programming algorithm is developed. • The algorithm is verified by comparing the results with two other dynamic programming algorithms. [ABSTRACT FROM AUTHOR]

A distribution case pack contains an assortment of varying quantities of different stock keeping units (SKUs) packed in a single box or pallet, with a goal of reducing handling requirements in the distribution chain. This article studies case pack procurement planning problems that address the trade-off between reduced order handling costs and higher inventory-related costs under dynamic, deterministic demand. The properties of optimal solutions for special cases of the problem involving one and two case packs are first established and these properties are used to solve the problem via dynamic programming. For the general model with multiple predefined case packs, which is shown to be strongly NP-hard, the exact approach is generalized to solve the problem in pseudopolynomial time for a fixed number of case packs. In addition, for large-size problems, the problem formulation is strengthed using valid inequalities and a family of heuristic solutions is designed. Computational tests show that these heuristic approaches perform very well compared to the commercial mixed-integer programming solver CPLEX. In addition to providing detailed methods for solving problems with deterministic demand, strategies for addressing problems with uncertain demands are discussed. [ABSTRACT FROM PUBLISHER]

Finding a reliable correspondence between two feature point sets efficiently and accurately is the first step in many computer vision applications. It is a crucial prerequisite for many remote sensings, target recognition, and photogrammetry applications. However, in different application scenarios, images may undergo different degrees of deformation, which makes the predefined geometric model no longer appropriate. This article attempts to remove mismatches between the set of assumed matching points. In order to achieve this goal, we propose an effective matching method, called "visualized local structure generation-Siamese attention" network (VLSG-SANet), which transform the task of eliminating mismatching of feature points into a dynamic visual similarity assessment. The VLSG-SANet can generate visual descriptors by using the local structure information about each feature point. We embed the attention module based on the Siamese network for the generated feature descriptors, allowing the network to adjust the descriptors dynamically. In order to prove the robustness and versatility of the VLSG-SANet algorithm, we have performed good experiments on open remote sensing data images and got good experimental results. [ABSTRACT FROM AUTHOR]