Search

Key semantics can come from everywhere on an image. Semantic alignment is a key part of few-shot learning but still remains challenging. In this paper, we design a Mixer-Based Semantic Spread (MBSS) algorithm that employs a {\it mixer} module to spread the key semantic on the whole image, so that one can directly compare the processed image pairs. We first adopt a convolutional neural network to extract features from both support and query images and separate each of them into multiple Local Descriptor-based Representations (LDRs). The LDRs are then fed into the {\it mixer} for semantic spread, where every LDR attracts complementary information from its peers. In this way, the objective semantic is made spread on the whole image in a data-driven manner. The overall pipeline is supervised by a voting-based loss, guaranteeing a good {\it mixer}. Visualization results validate the feasibility of our {\it mixer}. Comprehensive experiments on three benchmark datasets, miniImageNet, tieredImageNet, and CUB, show that our algorithm achieves the state-of-the-art performance in both $5$ -way $1$ -shot and $5$ -way $5$ -shot settings.

We focus on unsupervised representation learning for skeleton based action recognition. Existing unsupervised approaches usually learn action representations by motion prediction but they lack the ability to fully learn inherent semantic similarity. In this paper, we propose a novel framework named Prototypical Contrast and Reverse Prediction (PCRP) to address this challenge. Different from plain motion prediction, PCRP performs reverse motion prediction based on encoder-decoder structure to extract more discriminative temporal pattern, and derives action prototypes by clustering to explore the inherent action similarity within the action encoding. Specifically, we regard action prototypes as latent variables and formulate PCRP as an expectation-maximization (EM) task. PCRP iteratively runs (1) E-step as to determine the distribution of action prototypes by clustering action encoding from the encoder while estimating concentration around prototypes, and (2) M-step as optimizing the model by minimizing the proposed ProtoMAE loss, which helps simultaneously pull the action encoding closer to its assigned prototype by contrastive learning and perform reverse motion prediction task. Besides, the sorting can also serve as a temporal task similar as reverse prediction in the proposed framework. Extensive experiments on N-UCLA, NTU 60, and NTU 120 dataset present that PCRP outperforms main stream unsupervised methods and even achieves superior performance over many supervised methods.

Motivated by recent claims on the potential value of integration in warehouse management, this study evaluates the benefits arising from integrating the planning of order picking and packing processes in e-commerce warehouses. A set of research questions are proposed for exploring various benefits under different operational conditions and an experimental study is designed to answer them. In order to have a concrete model to represent the integrated planning method, a mixed-integer nonlinear programming model is developed, and then compared against a non-integrated variation. The experimental study makes the comparisons by analysing the collected empirical data from a real-life warehouse. Our findings indicate that integrated picking and packing planning can yield improved performance in different aspects under different configurations of objectives, order quantities, order categories or workforce allocation.

Time-lapse microscopy images generated by biological experiments have been widely used for observing target activities, such as the motion trajectories and survival states. Based on these observations, biologists can conclude experimental results or present new hypotheses for several biological applications, i.e. virus research or drug design. Many methods or tools have been proposed in the past to observe cell and particle activities, which are formed as single cell tracking and single particle tracking problems, by using algorithms and deep learning technologies. In this article, a review for these works is presented in order to summarize the past methods and research topics at first, then points out the problems raised by these works, and finally proposes future research directions. The contributions of this article will help researchers to understand past development trends and further propose innovative technologies.

This work focuses on image anomaly detection by leveraging only normal images in the training phase. Most previous methods tackle anomaly detection by reconstructing the input images with an autoencoder (AE)-based model, and an underlying assumption is that the reconstruction errors for the normal images are small, and those for the abnormal images are large. However, these AE-based methods, sometimes, even reconstruct the anomalies well; consequently, they are less sensitive to anomalies. To conquer this issue, we propose to reconstruct the image by leveraging the structure-texture correspondence. Specifically, we observe that, usually, for normal images, the texture can be inferred from its corresponding structure (e.g., the blood vessels in the fundus image and the structured anatomy in optical coherence tomography image), while it is hard to infer the texture from a destroyed structure for the abnormal images. Therefore, a structure-texture correspondence memory (STCM) module is proposed to reconstruct image texture from its structure, where a memory mechanism is used to characterize the mapping from the normal structure to its corresponding normal texture. As the correspondence between destroyed structure and texture cannot be characterized by the memory, the abnormal images would have a larger reconstruction error, facilitating anomaly detection. In this work, we utilize two kinds of complementary structures (i.e., the semantic structure with human-labeled category information and the low-level structure with abundant details), which are extracted by two structure extractors. The reconstructions from the two kinds of structures are fused together by a learned attention weight to get the final reconstructed image. We further feed the reconstructed image into the two aforementioned structure extractors to extract structures. On the one hand, constraining the consistency between the structures extracted from the original input and that from the reconstructed image would regularize the network training; on the other hand, the error between the structures extracted from the original input and that from the reconstructed image can also be used as a supplement measurement to identify the anomaly. Extensive experiments validate the effectiveness of our method for image anomaly detection on both industrial inspection images and medical images.

Please see the Editorial Comment by HaiThuy N. Nguyen discussing this article. Background: Use of upper gastrointestinal (UGI) series to diagnose UGI obstruction has raised concern in neonates and ...

As a raw material, applications of waste activated carbon (WAC) in synergy with coal powder to prepare coal water slurry (CWS) and perform high-temperature gasification is a novel strategy for disposing WAC. Combining BET surface analysis, contact angle measurement, and thermogravimetric-Fourier transform infrared (TG-FTIR) methods, the physico-chemical properties and CO2 gasification behaviors of coal-waste activated carbon-slurry (CWACS) were evaluated. Compared to coal, WAC exhibited an excellent adsorption capacity and a stronger surface wettability. When the WAC mixing ratio was increased from 0 to 15%, the CWS solid concentration decreased from 62.28% to 59.44%. CWACS exhibited a shear-thinning flow behavior, and the higher the WAC mixing ratio, the more obvious the pseudoplastic characteristics. Moreover, WAC enhanced the storage stability of the CWS. Under the catalysis of Na, Ca, and Fe among others, the CO2 gasification reaction of CWACS coke moved forward to the low-temperature stage. The high fixed carbon content of WAC, as well as its unique tar adsorption and catalytic cracking effects, increased the CO gas component in the synthesis gas.

Recent image-generation methods have demonstrated that realistic images can be produced from captions. Despite the promising results achieved, existing caption-based generation methods confront a dilemma. On the one hand, the image generator should be provided with sufficient details for realistic hallucination, meaning that longer sentences with rich content are preferred, but on the other hand, the generator is meanwhile fragile to long sentences due to their complex semantics and syntax like long-range dependencies and the combinatorial explosion of object visual features. Toward alleviating this dilemma, a novel approach is proposed in this article to hallucinate images from attribute pairs, which can be extracted from natural language processing (NLP) toolsets in the presence of complex semantics and syntax. Attribute pairs, therefore, enable our image generator to tackle long sentences handily and alleviate the combinatorial explosion, and at the same time, allow us to enlarge the training dataset and to produce hallucinations from randomly combined attribute pairs at ease. Experiments on widely used datasets demonstrate that the proposed approach yields results superior to the state of the art.

In large scale emergency scenarios, massive content for searching, asking for help, and rescue will be generated and transmitted in Intelligent Internet of Vehicular Things (IIoVT). However, IP-networks based emergency systems make rescue decisions on remote emergency centers, leading to in-efficient content dissemination and a high-latency response. Moreover, a few previous works address trust issues in the emergency systems, resulting in fake content and malicious emergency services. To address above challenges, we propose an emergent semantic-based information-centric fog system, which realizes trustworthy and intelligent emergency analysis and management. First, we design an efficient emergency content dissemination network for aggregating and analyzing emergency information. Besides, we propose a semantic-based trustworthy routing scheme that filters fake content from malicious entities. Moreover, we implement a real testbed and a simulator to evaluate the benefit and performance of the proposed system. The results show that the proposed system achieves a short average semantic analyzing time and a low failure rate of emergency services.

The development of medical imaging techniques has greatly supported clinical decision making. However, poor imaging quality, such as non-uniform illumination or imbalanced intensity, brings challenges for automated screening, analysis and diagnosis of diseases. Previously, bi-directional GANs (e.g., CycleGAN), have been proposed to improve the quality of input images without the requirement of paired images. However, these methods focus on global appearance, without imposing constraints on structure or illumination, which are essential features for medical image interpretation. In this paper, we propose a novel and versatile bi-directional GAN, named Structure and illumination constrained GAN (StillGAN), for medical image quality enhancement. Our StillGAN treats low- and high-quality images as two distinct domains, and introduces local structure and illumination constraints for learning both overall characteristics and local details. Extensive experiments on three medical image datasets (e.g., corneal confocal microscopy, retinal color fundus and endoscopy images) demonstrate that our method performs better than both conventional methods and other deep learning-based methods. In addition, we have investigated the impact of the proposed method on different medical image analysis and clinical tasks such as nerve segmentation, tortuosity grading, fovea localization and disease classification.

In this paper, the synthesis of 2,4,6-trinitrophenol from phenol was improved in the range of 0–65°C, and the yield was 79.4%. The effectiveness of the synthesis process was confirmed by Nuclear Ma...

As a component of servicing car body, the internal interfaces of aluminum alloy carbody include all connections of equipments hanged under floor and mounted on roof, which are expected to form the weak coupling relationship. For an imported prototype with primary hunting phenomenon, a dynamical design methodology of speeding-up bogies was proposed. The analysis graph of full-vehicle stability properties and variation patterns is used to clarify a self-adaptive improvement direction, i.e., λeN ≥ λemin, and λemin = (0.03–0.05). Therefore, the central hollow tread wear can be self-cleaned in time or regularly by crossing over the dedicated lines of different speed-grades. The modified strategy with strong/weak internal interface transaction of servicing car body was furthermore formulated based on the dynamical condensation method of component interface displacements. The causal relationship between bogie vibration alarm and car body fluttering phenomenon was then demonstrated by using techniques of rigid-flex coupling simulation. The self-excited vibration of traction converter intersects with the unstable hunting oscillation, ca. 9.2/9.3 Hz, which is consistent with the conclusions of tracking-test investigations on two car body fluttering formations. The technical space to promote the construction speed is thereby lost completely because of ride comfort decline, unsafe vibration of onboard electrical equipments, and weld fatigue damage of aluminum alloy car body. However, the rigid-flex coupling simulation analyses of trailer TC02/07 confirm that the safety threshold of bogie vibration warning can be appropriately increased as long as the lateral modal frequency of traction converters is greater than 12 Hz, preferably close to 14 Hz.

With the continuous development of social economy and the intensification of social competition, human resource management plays a more and more important role in the whole resource system. How to give full play to the advantages of human resources has become the key issue of human resource management evaluation. However, the current human resource management evaluation system has some problems, such as poor timeliness, one-sidedness, and subjectivity. Therefore, this paper proposes a BP image neural network optimized based on the simulated annealing algorithm to realize enterprise human resource management evaluation and image analysis. Through the learning of different time series samples, the average weight distribution scheme of main indicators is obtained, in which the average weight proportions of c1, c2, c3, and c4 are 25.5%, 24.8%, 17.9%, and 31.9%, respectively. In the comprehensive evaluation of enterprise employees, the error between the actual output and expected output is less than 4.5%. The results show that the BP image neural network based on simulated annealing algorithm has high accuracy in the image analysis and evaluation of enterprise human resource management. The output analysis results meet the actual needs of the enterprise and the personal development of employees and provide a decision-making scheme for the evaluation of enterprise human resource management.

The dynamic wireless charging system (DWCS) is developed to solve the problems of large battery volume and mileage anxiety of electric vehicles. However, the accurate position detection for electric vehicle DWCS is facing challenge. The traditional communication, detection and estimation methods are difficult to accurately obtain the position. To tackle this problem, the visual inertial simultaneous localization and mapping (VI-SLAM) method is applied to the electric vehicles DWCS. Firstly, the graph optimization based tight coupling method is used to integrate the monocular visual and inertial measurement unit (IMU) measurements. Secondly, the NVIDIA TX2 and MYNT VI-sensor suite are assembled, which the MTi300-IMU is treated as the ground truth system. Finally, the mobile vehicle is controlled to race on the simulated DWCS pathway. The experimental result shows that the method achieves great performance with the accuracy of centimeter level. In particular, the root mean square error (RMSE) of pose (i.e. position and orientation) in X, Y, Z directions are 0.086 m, 0.092 m, 0.102 m and 2.423°, 1.682°, 2.501°, respectively. Compared with the smooth variable structure filter (SVSF) based SLAM method, 0.106 m, 0.098 m, 0.130 m and 3.069°, 3.261°, 2.961°, the accuracy of ours is increased by 18.87%, 6.12%, 21.54% and 21.05%, 45.42%, 15.54%, respectively.

Encapsulation of microfiber Mach-Zehnder interferometer (MMZI) sensor for temperature and salinity sensing in seawater is developed. It includes a C-shape metal tube with a slit on the bottom and polymer adhesive with large elasticity modulus, by which a more robust sensor with relatively fast response can be obtained. To evaluate the performance of sensor after encapsulation, effects of different polymer adhesives on tension withstanding, temperature and salinity sensitivities are evaluated, and dependence of response time on shape of metal tube is investigated. Results show that sensitivities of the sensor in temperature and salinity sensing can reach to be about −1.77 nm/°C and 1.18 nm/ ‰, and response times are about 219.0 ms and 12.5 ms, respectively. By comparing temperatures and salinities measured by sensor with those measured by commercial device, accuracy of the sensor after encapsulation is proved. In addition, cross-sensitivities of strain/pressure and short/long-term stability after encapsulation are also evaluated. Finally, based on the basic structure, further improvements are attempted, such as double encapsulation with stainless steel mesh and complex encapsulation with two kinds of polymer. Encapsulations of MMZI demonstrated here show advantages of low cost, improved robustness, relatively fast response, long-term stability and easy expandability, which will not only afford a critical support for optical sensors operated in the ocean, but also offer some useful references for microfiber sensors used in other sensing applications.

Nitric oxide (NO)-mediated vasodilatation is a fundamental response of vasculature, however, regulation of NO signaling pathway on resistance vessels in the older adult with hypertension is still unclear. The 16-week-spontaneously hypertensive rats (SHR), the 18-month-SHR (OldSHR), and the age-matched Wistar-Kyoto rats were used to study the changes of mesenteric resistance artery dilatation caused by sodium nitroprusside (SNP). After pre-vasoconstriction by norepinephrine (NE), the response of endothelium-denuded mesenteric artery ring to SNP was observed, and the changes in vascular response after pharmacological interventions of key nodes in the NO/sGC/cGMP/PKG1α signaling pathway were observed as well. RNA sequencing and functional enrichment analyses were used to provide information for conducting validation experiments. Vasodilation of NO in OldSHR was decreased, which significantly correlated with the reduction of PKG-mediated effect. Functional enrichment analysis of RNA sequencing showed that genes encoding important proteins such as sGC and MYPT1 (protein phosphatase 1 regulatory subunit 12A) were downregulated in OldSHR. Molecular biology validation results showed that mRNA expression of both α and β subunits of sGC were reduced, while mRNA and protein expression of PKG1α were reduced in OldSHR. More importantly, the expression of MYPT1 and pS668-MYPT1 was significantly reduced in OldSHR, even under the treatment of SNP. The experiment also revealed an enhanced cAMP system in vasodilation in hypertension, while this function was completely lost in the OldSHR. Therefore, an NO-mediated decrease in vascular smooth muscle relaxation was found in the OldSHR. The dysfunction in cGMP-PKG signaling, in particular, decreased pS668-MYPT1 was mechanistically involved.

In order to improve the classification performance of the model for different kinds of arrhythmias, a parallel classification model of arrhythmia based on DenseNet-BiLSTM is researched and proposed. Firstly, the model adopts a parallel structure. After wavelet denoising and heartbeat segmentation of ECG signals, this model can simultaneously capture the waveform features of small-scale heartbeat and large-scale heartbeat containing RR interval; Then, based on deep learning, Densely connected convolutional network (DenseNet) is applied to improve the model's ability to extract local features of ECG signals, and bidirectional long short-term memory network (BiLSTM) is introduced to improve the performance of the model in extracting time series features of ECG signals; Finally, weighted cross entropy loss function is used to alleviate the class imbalance of arrhythmia, and Softmax function is applied to achieve 4 classifications of arrhythmia. Experiments based on MIT-BIH arrhythmia database show that under the intra-patient paradigm, training time for each epoch, overall accuracy, F 1 and specificity are 42 s, 99.44%, 95.89% and 99.32%, respectively; Under the inter-patient paradigm, training time for each epoch, overall accuracy, F 1 and specificity are 23 s, 92.37%, 63.49% and 94.51%, respectively. Compared with other classification models, the model proposed in this paper has a good classification effect and is expected to be used in clinical auxiliary diagnosis.

In the past decade, human activity recognition (HAR) has grown in popularity due to its applications in security and entertainment. As recent years have witnessed the emergence of health care and exoskeleton robotics which make use of wearable suits, human–machine interaction based on action recognition performs an important role in multimedia applications. Considering the limitations of the application scenario, the surface electromyography (sEMG) signal stands out in many wearable data collection devices for HAR. That is because: (1) timely feedback; (2) no damage to the human body; and (3) the wide range of recognizable actions. However, existing public datasets of sEMG contained relatively few activities, and several large-scale datasets only collected the action of the hand. In addition, the processing of sEMG signals is a new field with no effective evaluation system for it. To tackle these problems, we establish a novel dataset for HAR on lower-limb sEMG named “HAR-sEMG,” using 6 sEMG signal sensors attached to the left leg. A benchmark summarizing experiments with many combinations of existing high-dimensional signal processing algorithms-based manifold learning on our dataset is also provided for a performance analysis.

Action recognition via 3D skeleton data is an emerging important topic. Most existing methods rely on hand-crafted descriptors to recognize actions, or perform supervised action representation learning with massive labels. In this paper, we for the first time propose a contrastive action learning paradigm named AS-CAL that exploits different augmentations of unlabeled skeleton sequences to learn action representations in an unsupervised manner. Specifically, we first propose to contrast similarity between augmented instances of the input skeleton sequence, which are transformed with multiple novel augmentation strategies, to learn inherent action patterns (“pattern-invariance ”) in different skeleton transformations. Second, to encourage learning the pattern-invariance with more consistent action representations, we propose a momentum LSTM , which is implemented as the momentum-based moving average of LSTM based query encoder, to encode long-term action dynamics of the key sequence. Third, we introduce a queue to store the encoded keys, which allows flexibly reusing proceeding keys to build a consistent dictionary to facilitate contrastive learning. Last, we propose a novel representation named Contrastive Action Encoding (CAE) to represent human’s action effectively. Empirical evaluations show that our approach significantly outperforms hand-crafted methods by 10–50% Top-1 accuracy, and it can even achieve superior performance to many supervised learning methods (Our codes are available at https://github.com/Mikexu007/AS-CAL ).

BACKGROUND Tenofovir disoproxil fumarate (TDF) is a prodrug of a nucleotide analogue. As an antiviral drug, TDF has been proposed in the first-line treatment of chronic hepatitis B (CHB). Qingzhong, a brand name of TDF, commercialized by Jiangsu Chia-tai Tianqing Pharmaceutical Co Ltd., and Viread, another brand name of TDF, commercialized by GlaxoSmithKline, have both been approved by the State Food and Drug Administration, China. AIM To investigate the efficacy and safety of the two TDF agents in the treatment of Chinese CHB patients. METHODS This trial was registered at ClinicalTrials.gov with the identifier number of NCT02287857. A total of 330 Chinese CHB patients, among which 232 were hepatitis B e antigen (HBeAg)-positive, were included in this 5-year-long, multicenter, double-blinded, double-dummy, randomized-controlled, non-inferiority phase III trial. The participants were initially randomized into two groups: Group A (n = 161), in which the participants received 300 mg Qingzhong once a day for 48 wk; and Group B, in which the participants received 300 mg Viread once a day for 48 wk. Starting from week 49, all the participants in Groups A and B received 300 mg Qingzhong once a day until the 96th week. In this study, the primary endpoint was the decrease in plasma level of hepatitis B virus (HBV) DNA at the 96th week, while the secondary endpoints were suppression of HBV replication, alanine aminotransferase (ALT) normalization, HBeAg loss, and HBeAg seroconversion rates. RESULTS For the participants with HBeAg-positive CHB, the decrease in mean HBV DNA level relative to the baseline value was comparable between Groups A and B (5.77 vs 5.73 log10 IU/mL, P > 0.05) at the 96th week. In addition, similar percentages of HBeAg-positive participants in the two groups exhibited undetectable levels of HBV DNA, HBeAg loss, and HBeAg seroconversion (71.05% vs 77.97%, 31.00% vs 27.27%, and 20.22% vs 15.79%, respectively, in Group A vs Group B; P > 0.05). For the participants with HBeAg-negative CHB, the decrease in mean HBV DNA level relative to the baseline value was also comparable between Groups A and B (4.46 vs 4.70 log10 IU/mL, P > 0.05) at the 96th week. In addition, similar percentages of HBeAg-negative participants in the two groups exhibited undetectable levels of HBV DNA (87.23% vs 94.12% in Group A vs Group B, respectively; P > 0.05). Finally, similar percentages of CHB patients (HBeAg-positive or HBeAg-negative) in the two groups exhibited normalization of ALT (80.14% vs 84.57% in Group A vs Group B, respectively; P > 0.05), and similar incidences of adverse events were observed (106 vs 104 in Group A vs Group B, respectively; P > 0.05). CONCLUSION Both Qingzhong and Viread are effective and safe in the treatment of Chinese CHB patients according to the results of our clinical trial.

Extraperitoneal colostomy (EPC) after laparoscopic abdominoperineal resection (APR) remains a challenge for surgeons. This study aims to summarize our laparoscopic EPC method and assess its effects versus a transperitoneal colostomy (TPC) for patients with rectal tumors. A total of 133 patients with rectal cancer treated with laparoscopic APR between May 2009 and May 2020 were retrospectively reviewed. The clinical data, including demographics, comorbidities, tumor stage, colostomy duration, and complications were compared between the EPC group and the TPC group. The EPC group included 83 patients whose extraperitoneal tunnels were created using a cannula through a trocar port, and the TPC group included 50 patients whose stomata were formed traditionally. There were no differences in colostomy time [(23.1 ± 6) min vs. (21.4 ± 4) min, P = 0.078], number of parastomal dermatitis patients (5 vs. 2, P = 0.916), or number of stomal stenoses (1 vs. 1, P = 0.715) between the EPC and TPC groups. No cases of parastomal hernia developed in the EPC group, whereas 4 patients were diagnosed with a parastomal hernia; the difference between the two groups was statistically significant (P = 0.036). Laparoscopic EPC have a lower incidence of parastomal hernia than TPC. It is easy and inexpensive to create an extraperitoneal tunnel using a cannula through a trocar port.

Background Obesity is typically associated with metabolic dysfunction, but its impact on hepatocellular carcinoma (HCC) remains unclear in patients with chronic hepatitis B (CHB). Aim To study the effect of obesity on HCC development in patients with CHB receiving antiviral therapy. Methods We included patients from a Chinese multicentre, prospective, observational, treated CHB cohort in this study. General obesity was evaluated by body-mass index (BMI). Central obesity was evaluated by waist circumference, waist-to-hip ratio and waist-to-height ratio. Results A total of 5754 nucleos(t)ide analogue treated patients were enrolled in the analysis. The 5-year cumulative incidence of HCC was 2.9%. Waist-to-height ratio performed better in predicting HCC development than BMI, waist circumference or waist-to-hip ratio. Patients with central obesity (defined as waist-to-height ratio >0.5) had significantly higher 5-year incidence of HCC than those without central obesity in the overall population (3.9% vs 2.1%, hazard ratio [HR]: 2.06, P = 0.0001) and 745 propensity score matched pairs (4.7% vs 2.3%, HR: 2.04, P = 0.026), respectively. Besides cirrhosis status and aMAP HCC risk score, central obesity was also independently associated with HCC risk (HR: 1.63, P = 0.013). Waist-to-height ratio gain within 1 year was associated with a significantly higher HCC risk with an adjusted HR value of 1.88 (95% confidence interval: 1.12-3.13, P = 0.017). Conclusions Central obesity, evaluated by the waist-to-height ratio, was associated with a twofold increase in HCC risk among CHB patients receiving antiviral treatment, highlighting the important role of abnormal metabolic function in the progression of liver disease.

The TM02 mode has advantage in power capacity over the fundamental TM01 mode in relativistic backward-wave oscillator (RBWO), whereas in our formal experiment, the microwave pulse seemed not as wide as we expected. In this article, the power capacity of the TM02 mode is investigated theoretically and the mechanisms corresponding to the pulse shortening are analyzed in detail. It is found that the pulse shortening in the RBWO results from the breakdown on the window, the bombardment of the stray electrons traveling in the low guiding magnetic field, and the breakdown induced by the intensive electric field on the surface of the slow wave structure (SWS). The approach of eliminating the pulse shortening is put forward based on the theoretical research and the numerical optimization using our artificial intelligence (AI) code and is validated in the experiment. With larger aperture antenna adopted, post resonator simplified, and emitting electric field greatly decreased, the pulsewidth is lengthened from ~9 to 18 ns with the output power of 360 MW at guiding magnetic field of 0.9 T.

Background How to treat infantile hepatitis B virus (HBV) infection remains a controversial issue. The nucleoside analogue lamivudine (LAM) has been approved to treat children (2 to 17 years old) with chronic hepatitis B. Here, we aimed to investigate the benefit of LAM treatment in infantile hepatitis B. Case summary A 4-mo-old infant born to a hepatitis B surface antigen (HBsAg)-positive woman was found to be infected by HBV during a health checkup. Liver chemistry and HBV seromarker tests showed alanine aminotransferase of 106 U/L, HBsAg of 685.2 cut-off index, hepatitis B "e" antigen of 1454.0 cut-off index, and HBV DNA of > 1.0 × 109 IU/mL. LAM treatment (20 mg/d) was initiated, and after 19 mo, serum HBsAg was entirely cleared and hepatitis B surface antibody was present. The patient received LAM treatment for 2 years in total and has been followed for 3 years. During this period, serum hepatitis B surface antibody has been persistently positive, and serum HBV DNA was undetectable. Conclusion Early treatment of infantile hepatitis B with LAM could be safe and effective.

Background:Increased traumatic brain injury (TBI) risk was found in patients with bipolar disorder (BPD). Whether the medications for BPD and dosage moderate the risk of TBI is not clear.Aim:This study aimed to determine whether an association exists between BPD and TBI and whether the prescription of psychotropics moderates TBI risk.Methods:A total of 5606 individuals who had received diagnoses of BPD between January 1, 1997 and December 31, 2013 and 56,060 matched controls without BPD were identified from Taiwan’s National Health Insurance Research Database. Cases and controls were followed until the date of TBI diagnosis.Results:BPD was associated with a high risk of TBI (adjusted hazard ratio (aHR): 1.85; 95% CI: 1.62–2.11). Patients with BPD, with or without a history of psychiatric hospitalization, had increased risks of TBI (aHR: 1.94, 95% CI: 1.57–2.4 and aHR: 1.82, 95% CI: 1.55–2.1, respectively). The prescription of typical antipsychotics (0 Conclusion:Patients with BPD have a higher risk of TBI. The use of typical antipsychotics, TCAs, or high-dose BZDs increases the risk of TBI in BPD.

Depression is a common comorbid disorder associated with breast cancer, and it can have considerable physical and psychological impacts. Circulating cytokines have been proposed as a potential tool to predict depression in various diseases; however, limited studies have specifically examined it in breast cancer. In this study, we examined and compared the prediction ability of various circulating cytokines for depression in patients with breast cancer. Eighty-three patients with a new diagnosis of breast cancer not receiving chemotherapy were recruited; among them, 15 patients had depression and 68 did not have depression. Depression was evaluated using the Patient Health Questionnaire 9 (PHQ-9). Cytokine levels in the serum were measured using an immunology multiplex assay. Two types of cytokines were assayed: (1) proinflammatory cytokines (interleukin [IL]-1β, IL-2, IL-6, IL-12, IL-17A, interferon [IFN]γ, and tumor necrosis factor [TNF]α) and (2) anti-inflammatory cytokines (IL-4, IL-5, IL-10, and IL-13). Receiver operating characteristic (ROC) analysis was performed to calculate the area under the curves (AUCs), sensitivities, and specificities of circulating cytokines for predicting depression. As a result, IL-2 (AUC = 0.78) and IL-5 (AUC = 0.76) demonstrated good predictability for depression, even after controlling for the covariates (i.e. age, education, stage of cancer, surgery, radiation therapy, and hormone therapy). The optimal cut-off value of IL-2 for predicting depression was 1.06 pg/mL with a sensitivity of 86.7% and a specificity of 52.9%; this cytokine also had the best prediction ability in this study. Owing to the prediction ability and practical feasibility of circulating cytokines, they may be used as a valid laboratory diagnostic tool for depression in breast cancer.

Detecting relationships between objects is important for the complete understanding of visual scenes, which will be helpful for applications such as visual question answering, image search, and robotic interactions. It is however a challenging task due to the high variation of object appearance and interactions, and the often incomplete annotations. In this paper, we propose a fast method for visual relationship detection based on recurrent attention and negative sampling. First, to learn non-visual features, we use the Word2Vec model to extract semantic embedding features of object categories, and use binary masks to represent spatial location features. And we integrate the recurrent attention mechanism into the detection pipeline, enabling the network to focus on several specific parts of an image when scoring predicates for a given object pair. Then we use an undersampling technique to alleviate the influence of imbalanced annotations, particularly for zero-shot detection. The proposed method is simple but experiments prove that it is efficient and achieves state-of-the-art results on the benchmark VRD and Visual Genome (VG) datasets in most cases.

The multimodal based human action recognition is an attracting increasing topic since the different modalities can provide complementary information. However, it is difficult to improve the recognition performance due to the limitation of the ability to learn spatial-temporal features. In this paper, we propose a novel approach for multimodal human action recognition by learning complementary features from RGB-D sequence. Firstly, a segmented rank pooling method is proposed to compress the entire RGB-D sequence into dynamic images as inputs to the Convolutional Networks (ConvNets) for capturing spatial-temporal information. Secondly, a Segment Cooperative ConvNets (SC-ConvNets) is designed to learn the complementary features from RGB-D modalities. Different from the ConvNets-based approaches that learn multimodal features with multiple separate networks, the proposed SC-ConvNets enhance the recognition performance through joint optimization learning of single ConvNets. Then a single loss function is optimized to narrow the variance between RGB and depth modalities. In order to verify the effectiveness of the proposed method, we evaluate the SC-ConvNets on four public benchmark multimodal datasets, including NTU RGB+D 60, NTU RGB+D 120, SYSU 3D HOI, and PKU-MMD datasets. The proposed method achieves an accuracy of 89.4% and 91.2% for cross-subject and cross-view on NTU RGB+D 60, 86.9% and 87.7% for cross-subject and cross-setup on NTU RGB+D 120, 92.1% and 93.2% for cross-subject and cross-view on PKU-MMD, which are the state-of-the-art, and the accuracy of 84.2% and 82.9% for setting-1 and setting-2 on SYSU 3D HOI, which are comparable to the state-of-the-art. The experimental results demonstrate that the proposed segmented rank pooling can represent discriminative spatial-temporal information from the entire RGB and depth sequence, and the proposed SC-ConvNets can enhance recognition performance by learning complementary features from different modalities.

Recent studies have demonstrated that cigarette smoking is related to changes in brain structure and function. However, few studies focus on functional brain differences between male chronic smokers and nonsmokers in both local spontaneous activity and whole-brain functional networks. Our study recruited 67 chronic smokers and 43 nonsmokers who underwent functional magnetic resonance imaging (fMRI) scans to investigate functional activity and connectivity alterations in chronic smokers. We used the mean fractional amplitude of the low-frequency fluctuation (mfALFF) and mean regional homogeneity (mReHo) methods to investigate resting-state spontaneous activity in chronic smokers and nonsmokers. The graph theoretical analysis (GTA) and network-based statistical (NBS) analysis were also used to investigate functional connectivity alterations. Compared with nonsmokers, chronic smokers exhibited higher activation in the reward system and portions of the prefrontal cortex but lower activation in the default mode networks (DMN) and visual-related regions. In addition, correlation analysis was conducted to assess the associations between neuroimaging findings and the severity of nicotine dependence or expectations of smoking effects. Our results showed that certain brain regions correlated with the Fagerstrom Test for Nicotine Dependence (FTND), the positive aspect of the Drug Use Disorders Identification Test Extended (DUDIT-E), and the negative aspect of the DUDIT-E, especially in the attentional control networks and hippocampus. The graph theoretical analysis (GTA) results indicated chronic smokers exhibited a trend toward increased assortativity. Our network-based statistical (NBS) analysis revealed reduced functional connections between the subnetwork in the prefrontal cortex, olfactory cortex, angular gyrus, and cingulate gyrus of chronic smokers. We concluded that chronic smokers have neural adaptations in local spontaneous activity but remain healthy brain functional networks.

Recent research has devoted significant efforts on the utilization of WiFi signals to recognize various human activities. An individual’s limb motions in the WiFi coverage area could interfere with wireless signal propagation, that manifested as unique patterns for activity recognition. Existing approaches though yielding reasonable performance in certain cases, are ignorant of two major challenges. The performed activities of the individual normally have inconsistent speed in different situations and time. Besides that the wireless signal reflected by human bodies normally carries substantial information that is specific to that subject. The activity recognition model trained on a certain individual may not work well when being applied to predict another individual’s activities. Since only recording activities of limited subjects in a certain speed and scale, recent works commonly have a moderate amount of activity data for training the recognition model. The small-size data could often incur the overfitting issue that negative affect the traditional classification model. To address these challenges, we propose a WiFi-based human activity recognition system that synthesizes variant activities data through eight channel state information (CSI) transformation methods to mitigate the impact of activity inconsistency and subject-specific issues, and also design a novel deep-learning model that caters to the small-size WiFi activity data. We conduct extensive experiments and show synthetic data improve performance by up to 34.6% and our system achieves around 90% of accuracy with well robustness in adapting to small-size CSI data.

This paper presents a biologically-inspired saliency prediction method to imitate two main characteristics of the human perception process: focalization and orienting. The proposed network, named ACNet is composed of two modules. The first one is an essential concentrated module (CM), which assists the network to “see” images with appropriate receptive fields by perceiving rich multi-scale multi-receptive-field contexts of high-level features. The second is a parallel attention module (PAM), which explicitly guides the network to learn “what” and “where” is salient by simultaneously capturing global and local information with channel-wise and spatial attention mechanisms. These two modules compose the core component of the proposed method, named ACBlock, which is cascaded to progressively refine the inference of saliency estimation in a manner similar to that humans zoom in their lens to focus on the saliency. Experimental results on seven public datasets demonstrate that the proposed ACNet outperforms the state-of-the-art models without any prior knowledge or post-processing.

This paper presents fatigue behaviors and the stiffness degradation law of concrete continuous beams with external prestressed carbon fiber-reinforced polymer (CFRP) tendons. Three specimens were tested under fatigue loading, and the influence of different load levels on the stiffness degradation and fatigue life were studied, and it was found that the stiffness degradation of three test specimens exhibited a three-stage change rule, namely rapid decrease, stable degradation, and sharp decline, but there are obvious differences in the rate and amplitude of stiffness degradation. The load level has a significant influence on the fatigue life of the test specimens. An analytical model with load level considered was proposed to calculate the residual stiffness and predict the stiffness degradation, which is in good agreement with the test results. The model of stiffness degradation presents a possible solution for practical engineering applications of concrete continuous beams with externally prestressed CFRP tendons subjected to different fatigue loadings.

In this paper, through the application of 3D printing technology, insulators with different surface microstructures are prepared. Surface flashover test of the printed insulators shows that surface microstructures have a strong influence on the surface flashover voltage. This indicates that through proper surface microstructure design and the application of 3D printing technology insulators with high surface voltage withstand strength can be prepared. The influence of gas release characteristics, with and without thermal treatment, on the surface flashover voltages is studied. The results indicate that insulators with lower gas release rate own higher voltage withstand strength.

Our study aimed to investigate whether changes in brain function measured with functional magnetic resonance imaging (fMRI) can be detected among individuals with depressive disorders and suicidal ideation. The association between depression severity and brain images is also discussed. Our study recruited 111 participants in three groups: 35 depressive patients with suicidal ideation (SI), 32 depressive patients without suicidal ideation (NS), and 44 healthy controls (HCs). All participants were scanned using 3T MRI to obtain resting-state functional images, and functional connectivity (FC), amplitude of low-frequency fluctuation (ALFF), regional homogeneity (ReHo), and graph theoretical analysis (GTA) were performed. We found functional activity differences, such as the hippocampus and thalamus, in the SI group compared with the NS group. We also concluded lower activity in the thalamus and cuneus regions were related to suicidal ideation. We also found several functional connectivity of the brain areas, such as hippocampus, cuneus, and frontal regions, in the SI group correlated with Hamilton Depression Rating Scale (HAM-D) and Hospital Anxiety and Depression Scale (HADS). A graph theoretical analysis (GTA) and network-based statistical (NBS) analysis revealed different topological organization and slightly better local segregation of the brain network in healthy participants compared with those in depressive patients with suicidal ideation. We suggest that brain functional connectivity may be affected in depressive patients with suicidal ideation.

6-D pose estimation is an important branch in the field of vision measurement and is widely used in the fields of robotics, autonomous driving, and reality augmentation. The latest research trend in 6-D pose estimation is to train a deep neural network to directly predict the 2-D projection position of the 3-D keypoint from the image, establish the corresponding relationship, and, finally, use the perspective-n-point (PnP) algorithm to perform pose estimation. The current challenge of pose estimation is that, when objects are textureless, occluded, or scene-cluttered, the detection accuracy is reduced, and most of the existing algorithm models are large and cannot accommodate real-time requirements. In this article, we introduce a densely connected feature pyramid network (DFPN) that can efficiently integrate and utilize features. We combine the cross-stage partial network (CSPNet) with DFPN to design a new network for 6-D pose estimation, DFPN-6-D, a new approach for 6-D object pose estimation. DFPN-6-D can efficiently and accurately handle objects with textureless, occluded, and scene clutter and estimate their full 6-D poses in a single shot. Furthermore, we propose a new confidence calculation method and loss function for object pose estimation, which can fully consider spatial information. Finally, we propose a novel augmentation method for direct 6-D pose estimation approaches to improve performance and generalization ability in the case of occlusion, which is called 6-D augmentation. Our approach achieves a new state-of-the-art accuracy of 98.06 and 87.09 in terms of the ADD(-S) metric on the Linemod dataset and the Occluded-Linemod dataset, and our method also achieves the best result in terms of the different metric on the MULT-I dataset, the BIN-P dataset, and the T-LESS dataset, respectively, while still running end-to-end at over 65 frames/s. The experimental results demonstrate that our algorithm is robust to textureless materials and occlusion while running more efficiently than other methods. We also deploy our proposed method to a real robot to grasp and manipulate objects based on the estimated pose.

For a given software enhancement report, identifying its possible approval status could help software developers by suggesting feature enhancements to compete in the software industry. An automatic solution for the approval prediction of enhancements could assist all the participants in resolving enhancements. The key challenges are the preprocessing of noisy textual information and the state-of-the-art feature models to combine the syntactical and semantic word information available in the given text. To this end, we propose a deep learning based approach for the approval prediction of enhancement reports that incorporates the users’ sentiments involved in the text. First, we preprocess the textual information of all enhancement reports to avoid noise. Second, we compute the sentiment of each enhancement report using Senti4SD. Third, we combine the bag-of-words (BOW) representation and traditional word2vec based representation to learn the novel deep representation (a recurrent neural network (RNN) with attention based representation) of preprocessed text. Using an attention mechanism enables the model to remember the context over a long sequence of words in an enhancement report. Fourth, based on sentiment and deep representation, we train a deep learning based classifier for the approval prediction of enhancement reports. Finally, we reuse the 40, 000 enhancement reports from 10 real software applications to evaluate the proposed approach. The cross-application evaluation suggests that the proposed approach is accurate and outperforms the state-of-the-art. The results of the proposed approach improve the precision from 86.52% to 90.56%, recall from 66.45% to 80.10%, and f-measure from 78.12% to 85.01%.

Achieving high-efficiency photoelectrochemical water splitting requires a better understanding of ion kinetics, e.g., diffusion, adsorption and reactions, near the photoelectrode’s surface. However, with macroscopic three-dimensional electrodes, it is often difficult to disentangle the contributions of surface effects to the total photocurrent from that of various factors in the bulk. Here, we report a photoanode made from a InSe crystal monolayer that is encapsulated with monolayer graphene to ensure high stability. We choose InSe among other photoresponsive two-dimensional (2D) materials because of its unique properties of high mobility and strongly suppressing electron–hole pair recombination. Using the atomically thin electrodes, we obtained a photocurrent with a density >10 mA cm−2 at 1.23 V versus reversible hydrogen electrode, which is several orders of magnitude greater than other 2D photoelectrodes. In addition to the outstanding characteristics of InSe, we attribute the enhanced photocurrent to the strong coupling between the hydroxide ions and photo-generated holes near the anode surface. As a result, a persistent current even after illumination ceased was also observed due to the presence of ions trapped holes with suppressed electron-hole recombination. Our results provide atomically thin materials as a platform for investigating ion kinetics at the electrode surface and shed light on developing next-generation photoelectrodes with high efficiency., Achieving efficient photoelectrochemical cells requires insights into the ion kinetics at the surface of photoelectrode. Here, the authors show a new platform for investigating strong coupling between hydroxide ions and surface charge carriers by using monolayer InSe as the anode with high photocurrent density.

The existing relation extraction approaches select the relevant partial dependency structures and exhibit the limitation associated with long-distance dependencies. Moreover, due to the repeated use of the irrelevant redundant information and the lack of consideration of the key semantic details, the extraction of relations is relatively complex when the entities overlap. To address this limitation and effectively exploit the relevant information while ignoring irrelevant information, this paper proposes a simple but effective multistream semantics-guided dynamic aggregation graph convolution network (SG-DAGCN) to realize the extraction of overlapping relations. The proposed model constructs the entity relation graphs by enumerating the possible candidates and external auxiliary information and adaptively manages the relevant substructure. Subsequently, this framework models the relational graphs between the entities through a dynamic aggregation graph convolution module and gradually produces the discriminative embedded features and a refined graph through the dynamic aggregation of nodes. The proposed approach can effectively leverage the rich multiscale structural information and capture the long-distance dependencies between overlapping entities in long sentences. The results of the experiments conducted on two typical benchmark datasets show that the proposed model can achieve a high level of performance and outperform other state-of-the-art methods in both qualitative and quantitative aspects.

Volumetric representation is widely used in digital sculpting and terrain generation due to its highly modifiable characteristic. However, storing the entire raw voxel map requires a lot of memory, which limits its application on general machines. In this work, we propose to use a novel data structure based on the concept of layered depth-normal images to replace the commonly used scalar field for storing the volumetric information. The proposed data structure can be quickly edited, and the isosurface can be extracted by using general methods such as Marching Cubes and Dual Contouring. The proposed sculpting system can edit models with multiple resolution levels, and the surface between models of different resolutions can be connected seamlessly. Our method can achieve similar results as the existing volumetric methods while significantly reduces the memory usage and computation time. The model created by our system can be used in common game engines to make highly interactive games. Furthermore, all of our processes are parallel-friendly. We implemented a virtual reality digital content creation tool based on the proposed method to demonstrate the effectiveness and feasibility of our method.

The global population is at present suffering from a pandemic of Coronavirus disease 2019 (COVID-19), caused by the novel coronavirus Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2). The goal of this study was to use artificial intelligence (AI) to predict blueprints for designing universal vaccines against SARS-CoV-2, that contain a sufficiently broad repertoire of T-cell epitopes capable of providing coverage and protection across the global population. To help achieve these aims, we profiled the entire SARS-CoV-2 proteome across the most frequent 100 HLA-A, HLA-B and HLA-DR alleles in the human population, using host-infected cell surface antigen presentation and immunogenicity predictors from the NEC Immune Profiler suite of tools, and generated comprehensive epitope maps. We then used these epitope maps as input for a Monte Carlo simulation designed to identify statistically significant “epitope hotspot” regions in the virus that are most likely to be immunogenic across a broad spectrum of HLA types. We then removed epitope hotspots that shared significant homology with proteins in the human proteome to reduce the chance of inducing off-target autoimmune responses. We also analyzed the antigen presentation and immunogenic landscape of all the nonsynonymous mutations across 3,400 different sequences of the virus, to identify a trend whereby SARS-COV-2 mutations are predicted to have reduced potential to be presented by host-infected cells, and consequently detected by the host immune system. A sequence conservation analysis then removed epitope hotspots that occurred in less-conserved regions of the viral proteome. Finally, we used a database of the HLA haplotypes of approximately 22,000 individuals to develop a “digital twin” type simulation to model how effective different combinations of hotspots would work in a diverse human population; the approach identified an optimal constellation of epitope hotspots that could provide maximum coverage in the global population. By combining the antigen presentation to the infected-host cell surface and immunogenicity predictions of the NEC Immune Profiler with a robust Monte Carlo and digital twin simulation, we have profiled the entire SARS-CoV-2 proteome and identified a subset of epitope hotspots that could be harnessed in a vaccine formulation to provide a broad coverage across the global population.

Objective To investigate the effectiveness of arthroscopic surgery for combined tears of the posterior horn of the medial meniscus (PHMM) and the anterior horn of the lateral meniscus (AHLM). Methods Between September 2009 and December 2013, a retrospective investigation was performed on 48 patients (48 knees) with combined tears of PHMM and AHLM who underwent arthroscopic surgery. All patients underwent magnetic resonance imaging (MRI) diagnosis in the outpatient department. After admission, other basic examinations were performed. PHMM was treated with partial meniscectomy and AHLM was treated with external–internal suture or partial meniscectomy. Curative effect was evaluated according to Lysholm score and IKDC score. The differences of the functions as well as Lysholm and IKDC scores pre‐ and postoperatively were compared, while the diagnostic accuracy of MRI and arthroscopy for PHMM and AHLM were compared. Results Seven patients with combined tears of PHMM and AHLM were misdiagnosed during MRI examination when arthroscopy was used as the gold standard. All patients were followed up for an average of 34.58 months (range 10–52 months) and all incisions healed by first intention with no obvious complication. The preoperative and postoperative Lysholm scores were 47.22 ± 2.77 and 87.36 ± 5.45, respectively. The IKDC scores are 54.73 ± 4.65 preoperatively and 89.62 ± 3.71 postoperatively. The positive rates of the diagnosis through MRI and arthroscopic surgery is 85.42% and 100%, respectively. At the last follow‐up, the patients had no pain, weakness, and instability, and tenderness in medial and lateral joint space disappeared. Mcmurray test was weakly positive in four patients. Excellent outcome was achieved in 39 cases, and a good outcome was achieved in five cases; the good to excellent rate was 91.67%. Conclusion MRI examination of combined tears of PHMM and AHLM may result in misdiagnosis. Arthroscopic primary repair seemed to be an effective surgical option for treatment of combined tears of PHMM and AHLM., Schematic diagram of arthroscopy of meniscus injury following surgery; (A) The anatomic structure of the knee; (B) AHLM and PHMM injuries of the meniscus; (C) Description of the performed meniscus injury with suture anterior horn of the lateral meniscus; (D) Description of the performed meniscus injury with removal of the incomplete bacterial horn of the medial meniscus.

Background To determine the safety and efficacy of different doses of tolvaptan for treating Chinese cirrhotic patients with or without hyponatraemia who still had ascites after routine therapy with diuretics. Methods In the present placebo-controlled, randomized, double-blinded, multicentre clinical trial, patients with cirrhotic ascites who failed to adequately respond to a combination of an aldosterone antagonist plus an orally administered loop diuretic were randomly placed at a 4:2:1 ratio into 3 groups [the 15 mg/day tolvaptan group (N = 301), 7.5 mg/day tolvaptan group (N = 153) and placebo group (N = 76)] for 7 days of treatment. The effects and safety were evaluated on days 4 and 7. A change in body weight from baseline on day 7 of treatment was the primary endpoint. Results The administration of 7.5 or 15 mg/day tolvaptan significantly decreased body weight from baseline on day 7 of treatment compared to that with placebo treatment (P = 0.026; P = 0.001). For the secondary endpoints, changes in abdominal circumference from baseline and improvements in ascites were markedly different in the treatment groups and the placebo group on day 7 (P7.5 = 0.05, P15.0 = 0.002 and P7.5 = 0.037, P15.0 = 0.003), but there was no difference between the 7.5 mg/day and 15 mg/day dosage groups. The 24-h cumulative urine volume was higher in the 7.5 mg/day and 15 mg/day tolvaptan groups than the placebo group (P = 0.002, P P = 0.004). Sodium serum concentrations were higher in patients with hyponatraemia after tolvaptan treatment, with no significant difference between the two dosage groups. The incidence of serious adverse drug reactions was not different between the groups (P = 0.543). Conclusions Tolvaptan treatment at 7.5 mg per day might be a good therapeutic choice for Chinese cirrhotic patients with ascites who did not achieve satisfactory clinical responses to previous treatment regimens with combination therapy with an aldosterone antagonist and an orally administered loop diuretic. Trial registration NCT01349348. Retrospectively registered May 2011.

Previous findings regarding declines in cognitive functioning among patients with breast cancer (BC) before and after chemotherapy have been inconsistent. The present study explored the effect of BC and cancer-related chemotherapies on cognitive functioning. A cross-sectional design was adopted to compare BC patients before their chemotherapy treatment, BC patients 3 ~ 9 months after the completion of chemotherapy, and noncancer controls. Evaluations of cognitive functioning included subjective and objective dimensions, with focus on memory, executive functioning, attention, and language. ANCOVA and Pearson’s correlation analysis were used to examine the relationship among cancer, chemotherapy, cognitive performance, and psychological distress. After adjustment for intelligence quotient, anxiety, and depression, we found significant differences in the Semantic Association of Verbal Fluency between post-chemotherapy (C/T) patients and noncancer controls. Specifically, post-C/T patients scored lower than controls (p = 0.03, η2 = 0.07). No significant differences were found in other objective cognitive measures. However, both subjective and objective cognitive scores were significantly associated with depression, anxiety, and fatigue. In BC patients, levels of anxiety were positively correlated with measures of executive function. Among pre-C/T patients, self-perceived interference by fatigue was positively associated with better performances in some of the objective cognitive measures. Our findings suggest cognitive impairments in the domain of executive functioning among patients with BC who received chemotherapy. Providing relevant suggestions or strategies of managements for these negative consequences may help increase the long-term quality of life of patients with BC.