Search

Benefiting from the pay-as-you-go business model, cloud computing has significantly promoted service computing techniques in real-world industrial applications. Software applications based on cloud computing are becoming more and more popular. By integrating existing component cloud services through the internet, composite cloud systems can be built to meet sophisticated application logic. Stable execution of such systems is desirable in the long term so that the service-level agreements (SLAs), as well as users’ quality of experience (QoE), can be fulfilled. To achieve this goal, it is critical to identify and fault-tolerate system components at high risks of failing. This is extremely challenging due to the dynamic and uncertainty of the cloud environment that hosts the component cloud services. Nevertheless, existing approaches pay little attention to the modeling and analysis of system components’ reliability time series. To address the above issues, we first present a reliability evaluation method for component cloud services based on the reliability model and their failure probability under continuous client-side invocation tests. Then, we propose a perturbation-aware reliability sensitivity measurement approach (named PARS) for measuring the reliability sensitivity of component cloud services. It first analyzes the negative perturbations in component cloud services’ historical reliability time series based on the Markov chain rule. Then, it calculates the reliability sensitivity of component cloud services by analyzing how their reliability perturbations impact the reliability of the entire cloud system. To guarantee the execution quality of the composite cloud system, we further propose a proactive adaptation approach named PA-PARS that enables 1-out-of-2 N-version Programming fault-tolerance for composite cloud systems based on PARS. PA-PARS takes the reliability sensitivity of component cloud services estimated by PARS as input to assure the reliability of the cloud system. It consists of four parts: 1) risky system component identification; 2) adaptation trigger; 3) candidate component cloud service selection; and 4) NVP-based system construction as the proactive adaptation for the composite cloud system. The results of experiments conducted on two widely-used datasets demonstrate the effectiveness and efficiency of the proposed approaches in ensuring the reliability of composite cloud systems.

Here a 1×5 microfluidic optical switch using double drives is proposed. It has a simple structure and ignorable polarization dependent loss. First, the switch with the straight waveguide structure ...

The p-n homojunction devices based on 2-D-layered materials (2DLMs) with unusual physical characteristics are gaining extensive research on their use in photodetectors. However, the current fabrication of p-n homojunction devices based on 2DLMs needs a complex doping method. In our work, the laser-irradiation method is a simple and straightforward selective p-type doping technique for the 2-D semiconductor molybdenum ditelluride (MoTe2). The laser-oxidized product MoO x is responsible for p-type doping in MoTe2. The laser-irradiation technique has been used to directly write an in-plane MoTe2 p-n homojunction device, where the laser-scanned region is p-doped and the neighboring region is electrically n-doped by applying a positive back-gate voltage. The MoTe2 p-n homojunction device has been used as a photodetector, which shows a high photoswitching ratio of 105 under the illumination of 633 nm at $6.5~\mu \text{W}$ and a short photoresponse time of $232~\mu \text{s}$ . The selective p-doped technique using a laser-irradiation method can be used in future logic circuits based on 2-D materials.

The outstanding physical and electrical properties of transition metal dichalcogenides (TMDs) as semiconductor materials demonstrate a promising platform for future electronic devices. Among all the TMDs, MoTe2, in which the bandgap is close to that of the silicon (Si), is a more favorable candidate than others to be applied in next-generation integrated circuits (ICs). However, the conventional physical or chemical doping method is complicated for fabricating the MoTe2 logic ICs. The transistors with additional polarity gates (PGs) are defined as polarity-controllable transistors (PCTs). The PG can dynamically control the type of charge carriers (n- or p-type) in the source/drain by electrostatic doping without the need of any physical or chemical doping, and thus reconfigure the transistor between n-type and p-type. In our work, the ambipolar conduction property in MoTe2 enables the fabrication of high-quality polarity-controllable MoTe2 transistors (PCMTs) that are promising as building blocks to construct the MoTe2 logic ICs. The on/off ratios of the PCMTs are above 107 for both n-type and p-type. The highest field-effect mobility $\mu $ of p- and n-type MoTe2 transistors are 38 and 42, respectively. The inverter (INV) based on the PCMTs has achieved a high gain of 37. Furthermore, the logic-gate cell library, which includes INV, negative- AND (NAND), negative- OR (NOR), exclusive- OR (XOR), and maJority (MAJ) is demonstrated using PCMTs. The above-mentioned desirable properties make PCMTs promising for future applications in 2-D-semiconductor-material-based logic ICs.

Background Data on treatable traits (TTs) in different populations are limited. Objective To assess TTs in elderly patients with asthma and compare them to younger patients, to evaluate the association of TTs with future exacerbations, and to develop an exacerbation prediction model. Methods We consecutively recruited 521 participants at West China Hospital, Sichuan University based on the Australasian Severe Asthma Network, classified as elderly (n = 62) and nonelderly (n = 459). Participants underwent a multidimensional assessment to characterize the TTs and were then followed up for 12 months. TTs and their relationship with future exacerbations were described. Based on the TTs and asthma control levels, an exacerbation prediction model was developed, and the overall performance was externally validated in an independent cohort. Results A total of 38 TTs were assessed. Elderly patients with asthma had more chronic metabolic diseases, fixed airflow limitation, emphysema, and neutrophilic inflammation, whereas nonelderly patients with asthma exhibited more allergic characteristics and psychiatric diseases. Nine traits were associated with increased future exacerbations, of which exacerbation prone, upper respiratory infection–induced asthma attack, cardiovascular disease, diabetes, and depression were the strongest. A model including exacerbation prone, psychiatric disease, cardiovascular disease, upper respiratory infection–induced asthma attack, noneosinophilic inflammation, cachexia, food allergy, and asthma control was developed to predict exacerbation risk and showed good performance. Conclusions TTs can be systematically assessed in elderly patients with asthma, some of which are associated with future exacerbations, proving their clinical utility of evaluating them. A model based on TTs can be used to predict exacerbation risk in people with asthma.

The portability and low cost of e-commerce platforms have attracted more and more businesses and consumers. However, due to the virtual nature of e-commerce platforms, there is a serious information asymmetry between merchants and consumers–resulting in a serious credit problem of e-commerce platforms. The trust crisis remains a challenging issue for the development of e-commerce. Existing researches propose to strengthen trust supervision to protect the consumers. However, the trust crisis often hurts multiple parties’ interests. We can neither attribute the cause of dishonest transactions to a single party, nor supervise only a single party. To solve the problem of commercial breach of trust, we propose a margin policy-based consumer-to-business-to-consumer (or MP-C2B2C) model for e-commerce platforms. First, we add a third-party platform supervision mechanism to the C2C e-commerce model by presenting different margin policies. Second, we give the unit price trading mode, dual price trading mode and consumer reporting behavior mode under the margin policies based on two-way games. We investigate the game strategies of each participant in the e-commerce transaction process under MP-C2B2C model by comparing them with different representative game models. The research results show that participants in the transaction will be influenced by the margin system to choose the game strategy with better trustworthiness. Trustworthiness of the e-commerce platform is enhanced under the MP-C2B2C model.

The fall armyworm (FAW), Spodoptera frugiperda (J. E. Smith), is native to the Americas. It has rapidly invaded 47 African countries and 18 Asian countries since the first detection of invasion into Nigeria and Ghana in 2016. It is regarded as a super pest based on its host range (at least 353 host plants), its inherent ability to survive in a wide range of habitats, its strong migration ability, high fecundity, rapid development of resistance to insecticides/viruses and its gluttonous characteristics. The inherently superior biological characteristics of FAW contribute to its invasiveness. Integrated pest management (IPM) of FAW has relied on multiple applications of monitoring and scouting, agricultural control, chemical pesticides, viral insecticides, sex attractants, bio-control agents (parasitoids, predators and entomopathogens) and botanicals. Knowledge gaps remain to be filled to: (1) understand the invasive mechanisms of S. frugiperda; (2) understand how to prevent its further spread and (3) provide better management strategies. This review summarizes the biological characters of FAW, their association with its invasiveness and IPM strategies, which may provide further insights for future management.

A new image encryption scheme is proposed with a combination of block scrambling, closed-loop diffusion, and DNA molecular mutation. The new chaotic block scrambling mechanism is put forward to replace the traditional swapping rule by combining the rectangular-ambulatory-plane cyclic shift with the bidirectional random disorganization. The closed-loop diffusion strategy is designed to form a feedback system, which improves the anti-interference capacity of the algorithm. To further destroy the blocks characteristics and eliminate the correlations among adjacent blocks, two efficient methods of DNA molecular mutation are adopted in the mutation stage. Moreover, the proposed algorithm possesses a large key space and the keys are highly related with the plaintext image. Experimental results demonstrate that the suggested image encryption strategy is practicable and has strong ability against a variety of common attacks.

In perovskite solar cells (PSCs), structure evolution that can retard the interfacial charge recombination or minimize the optical loss is of crucial importance for further efficiency development but still poses substantial challenges. Here, a novel quasi-point-contact and rear-reflection (PCRR) structure is developed for efficient PSCs by inserting a discontinuous dielectric interlayer between the perovskite film and hole transport layer (HTL). In this architecture, a large portion of perovskite/HTL contact areas, particularly the defect enriched grain boundaries, are separated by the insulating interlayer while some discrete point-contacts are reserved at the convex of some perovskite grains for hole transport. The point contact leads to significantly reduced perovskite/HTL contact areas and thus remarkably restrained interfacial charge recombination. Moreover, the polysilane with a smaller refractive index coated on the perovskite layer leads to a significant internal light reflection at the polysilane/perovskite interface. Due to the simultaneously reduced charge recombination loss and enhanced light-harvesting, the resulting PSC yields a remarkable efficiency increment from 20.96% to 22.49%. In addition, the humidity stability of the device is also remarkably improved due to the in situ encapsulation of the perovskite film by the cross-linked polysilane layer. This study paves a brand new way of interface engineering for efficient PSCs.

This paper describes our approach for the Chinese clinical named entity recognition (CNER) task organized by the 2020 China Conference on Knowledge Graph and Semantic Computing (CCKS) competition. In this task, we need to identify the entity boundary and category labels of six entities from Chinese electronic medical record (EMR). We constructed a hybrid system composed of a semi-supervised noisy label learning model based on adversarial training and a rule post-processing module. The core idea of the hybrid system is to reduce the impact of data noise by optimizing the model results. Besides, we used post-processing rules to correct three cases of redundant labeling, missing labeling, and wrong labeling in the model prediction results. Our method proposed in this paper achieved strict criteria of 0.9156 and relax criteria of 0.9660 on the final test set, ranking first.

Robust superhydrophobic cotton fabrics with ultraviolet (UV)-proof performance were prepared via a simple one-pot method with composite coating of hierarchical CuS microflower nanoparticles and silicone elastomer polydimethylsiloxane (PDMS). The facile preparation of CuS microflower nanoparticles was conducted via relatively low-temperature reaction. The surface morphology, chemical composition, thermal degradation behavior, and crystal structure of the products were evaluated. The simultaneous roughening due to hierarchical rough morphology of CuS microflower nanoparticles and PDMS hydrophobization resulted in the coated fabric with microscale fibers exhibiting superhydrophobic property and its water contact angle was 159.3° ± 0.6°. The superhydrophobic textile also displayed outstanding UV-proof performance with a large ultraviolet protection factor of 535.36 due to ultraviolet-absorption, light scattering and frequent light reflection of hierarchical CuS nanoparticles. The superhydrophobic textile showed excellent stain resistance and self-cleaning property and also was able to separate oil and water with high efficiency and good separation reusability caused by the contrasting wettability of the coated textile for water droplets and oil droplets. The as-prepared superhydrophobic fabric displayed excellent robustness against 2000 cycles rubbing, water jetting and long-time UV illumination and satisfactory chemical stability. This superhydrophobic coating showed no negative impact on the mechanical strength of cotton fabric. The facile approach in favor of large-scale preparation showed promising potential for wide application of flexible superhydrophobic cotton textiles with ultraviolet-blocking performance.

We present an interesting transient effect found in silicon-on-insulator (SOI) transistors, where the drain current responds slowly to a back-gate voltage ( ${V}_{\text {BG}}$ ) pulse. This transient mechanism is due to the deep-depletion region in the SOI substrate induced by the ${V}_{\text {BG}}$ pulse and has been validated by experimental and TCAD simulation results. The deep-depletion characteristic can be employed for photodetection in a fully depleted (FD)-SOI MOSFET device. We have measured the photoresponse of the FD-SOI MOSFET under various light illumination conditions and studied systematically the impact of the ${V}_{\text {BG}}$ pulse duty ratio, amplitude, and period on the responsivity and detection range. An optimized device structure, fabricated in an advanced ultrathin body and buried oxide (BOX) (UTBB) SOI technology, shows low operation voltage and extended detection range.

OBJECTIVE To evaluate the efficacies of conventional ultrasound (US), US elasticity imaging (EI), and acoustic radiation force impulse (ARFI) elastography in breast malignancy diagnosis. METHODS We included 315 women (mean age, 44 years; range, 18-81 years) with 336 pathologically proven breast lesions in this retrospective study. All lesions underwent conventional US, EI, and ARFI (including virtual touch tissue imaging [VTI], virtual touch tissue quantification [VTQ], and virtual touch tissue imaging and quantification [VTIQ]) elastography. Multivariate logistic regression analysis was performed to assess 12 independent variables for malignancy prediction. Diagnostic performance was evaluated with receiver operating characteristic (ROC) curve analysis. RESULTS Irregular lesion shape was the strongest independent predictor for breast malignancy, followed by poorly defined margins, taller than wide dimensions, posterior echo attention, VTIQ, and VTI boundaries (P

Objective To investigate the association between non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio and degree of coronary artery stenosis proven by coronary angiography. Methods A total of 1867 patients were enrolled into this study and analyzed retrospectively. Three hundred eighty-five non-coronary artery disease hospitalized patients were selected as control group, 1482 patients diagnosed as coronary artery disease were classified into three subgroups according to the tertiles of their SYNTAX score. We compared the level of non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio among the three subgroups. The Spearman correlation was used to analyze the correlation between non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio and SYNTAX, logistic regression was used for analyzing independent predictors of coronary artery disease. Results The level of non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio was higher in coronary artery disease group compared with non-coronary artery disease group (P < 0.01). The Spearman correlation analysis showed that non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio were significantly correlated with SYNTAX score (r = 0.081, P < 0.001; r = 0.216, P < 0.001). In multivariate logistic regression analysis showed that non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio were independent predictors of coronary artery disease (odds ratio = 3.645, 95% confidence interval, 1.267–10.486; OR = 2.096, 95% confidence interval, 1.438–3.054). Conclusion Non–high-density lipoprotein cholesterol/apolipoprotein A-I and monocyte/high-density lipoprotein cholesterol ratio were associated with the severity of coronary artery lesions, which can be used as a biomarker for the evaluation of severity of coronary artery disease.

The thermal conductivity of two-dimensional materials like graphene can efficiently be tuned by introducing holes, in which the density and distribution of the holes are the key parameters. Furthermore, the distribution of holes can induce a variation as high as 74% in the thermal conductivity for porous graphene with a given density of holes. Therefore, an existing challenge is to find the optimal distribution of holes that can minimize or maximize the thermal conductivity of porous graphene as the design space expands dramatically with increasing hole density. We therefore apply an inverse design methodology based on machine learning to reveal the relationship between hole distribution and thermal conductivity reduction in monolayer graphene. The methodology reveals that holes that are randomly distributed transverse to the direction of heat flow, but that exhibit some periodicity along the direction of heat flow, represent the optimal distribution to minimizing the thermal conductivity for porous graphene. Lattice dynamics calculations and wave packet simulations reveal that this spatial distribution effectively causes localization of the phonon modes in porous graphene, which reduces the thermal conductivity. Overall, this work demonstrates the power of machine learning-based design approaches to efficiently obtain new physical insights for scientific problems of interest.

Coronary artery calcification affects the arteries that supply the heart with blood, and percutaneous coronary intervention (PCI) is a direct and effective surgery to alleviate this symptom. In this paper, we propose a framework to judge if a patient requires surgery, based on cardiac computerized tomography scans. We adopt generative adversarial network to segment the calcified areas from slices. This architecture provides an environment for the generator to perform joint learning from ground truth images and the high-resolution discriminator. We use images reconstructed using two types of filters to test our method. An F1 score of 96.1% and 85.0% was achieved for the soft and sharp filters. In addition, we explored different recurrent neural networks for making the final decision. Including long short-term memory, which was ultimately used to deal with the calcium score normalized by the age and score threshold. Using the soft reconstruction image as the input, the whole framework achieved an accuracy of 76.6%. These results certify that our method can precisely locate lesion in artery, and make a reasonable risk assessment for PCI.

We demonstrate dual-gated (DG) MoS2 field effect transistors (FETs) in which the degraded switching performance of multilayer MoS2 can be compensated by the DG structure. It produces large current ...

Electrochemical measurements at the single entity level provide ultra-sensitive tools for precise diagnosis and understanding of basic biological and chemical processes. By decoding current signatures, the single-entity electrochemistry provides abundant information on charges, sizes, shapes, catalytic performances and compositions. The accuracy of single-entity electrochemistry highly relies on the advanced instrumentation to achieve the amperometric resolution at sub-picoampere level and the temporal resolution at sub-microsecond level. Currently, it is still a challenge for paralleling amplifiers to allow the low-noise and high bandwidth single-entity electrochemical measurements. Herein, we developed a low-noise four-channel electrochemical instrumentation that integrates an Au electrode array with amplifiers in the circuit board. With this amplifier array, we achieved a high bandwidth (> 100 kHz) electrochemical measurements. The further practical experiments proved the capability of this amplifier array system in acquiring transient signals from both single-molecule detection with an Aerolysin nanopore, and single Pt nanoparticle catalysis during the dynamic collision process. Paired with appropriate microfluidic array systems, our instrumentation will enable an extraordinarily high-throughput feature for single-entity sensing.

Background：The prevalence of metabolic syndrome in the elderly is gradually increasing，which accounts for the largest burden of non-communicable diseases worldwide and has direct effects on health. Research on the relationship between uric acid and metabolic syndrome in the elderly is relatively lacking. The purpose of this study is to explore the diagnostic value of uric acid levels for metabolic syndrome, compared to other components of metabolic syndrome.Materials and methods：We collected the physical examination data of 1,267 elderly people in the community in Wuhan, and used SPSS IBM 22.0 for data analysis. Perform correlation analysis, logistic regression analysis and draw ROC curve. Results：The prevalence of hyperuricemia was 28.1%, and metabolic syndrome was 18.6%; the uric acid level of the non-metabolic syndrome group was lower than that of the metabolic syndrome group (337.31 vs 381.91 µmol/ L; PConclusion：We confirmed the correlation between uric acid levels and metabolic syndrome in the elderly Chinese population.

A completely non-blocking M × N electrically controlled optofluidic matrix switch that uses a 1 × 3 optical switch with a V-shaped microchannel as the switching unit is proposed. Its light paths and output ports are selected by a micro-actuator matrix and a control circuit. There are few reports of optofluidic matrix switches. Here the given electrostatic micro-actuator and the basic switch structure provide an effective feasible manner for the matrix switch due to the simple and compact structure as well as the operation style. The impacts of microchannels and intersecting waveguides on the switch performance are discussed, and multiple optimization schemes are proposed to reduce the insertion loss efficiently and significantly. The research results indicate that the M × N matrix switch has the advantages of good matrix controllability, simple structure, wide waveband (400–1700 nm), negligible polarization-dependent loss, small insertion loss, and low cross talk. For 1550 nm wavelength, the insertion loss of a 2 × 6 matrix switch is about 0.17–0.55 dB, and the maximum cross talk is less than − 26.8 d B . In addition, the performance parameters of a 4 × 8 matrix switch are given and compared with other reported matrix switches. The proposed M × N matrix switch solves the problem of large insertion loss of general optical matrix switches and can be expanded to a large-scale matrix switch. Moreover, the design of multiple output ports has more flexible applications in systems with multiple branch optical paths and network nodes.

Background Nonalcoholic fatty liver disease (NAFLD) and gallstone disease (GSD) often coexist in the general population owing to shared risk factors. This study explored the relationship between NAFLD and GSD in patients with type 2 diabetes. Methods We conducted a retrospective cross-sectional analysis of 4325 patients with type 2 diabetes. GSD and NAFLD were confirmed using ultrasonography. GSD was defined as either asymptomatic gallstones or previous cholecystectomy, and each was analyzed separately. Result There was no significant difference in the prevalence of GSD between patients with and without NAFLD (23.8% vs. 21.2%, P = 0.15). After case–control matching (1:1) of baseline data such as age, sex, duration of diabetes, and HbA1c between patients with and without NAFLD, there was still no significant difference in the prevalence of GSD (25.5% vs. 23.6%, P = 0.15). The prevalence of NAFLD in patients with asymptomatic gallstones was lower than that of patients without GSD (38.6% vs. 47.3%, P P P P P = 0.002) and cholecystectomy (OR = 1.946, 95%CI: 1.546–2.445, P < 0.001) were both strongly associated with NAFLD. However, asymptomatic gallstone (OR = 0.663, 95%CI: 0.513–0.856, P = 0.002) seemed to be negatively correlated with NAFLD. Conclusions The prevalence of GSD was similar in patients with type 2 diabetes with and without NAFLD. The higher proportion of cholecystectomy and lower proportion of asymptomatic gallstones in patients with NAFLD suggests that NAFLD may increase the risk of complications of GSD.

In this work, we present a brand-new narrowband red-emitting fluoroperovskite via the introduction of Mn4+ into NaZnF3 through a facile co-precipitation method at room temperature. The physicochemical properties of the fluoroperovskite such as crystal and electronic structures, morphology, and elemental composition, as well as its spectroscopic properties such as luminescence behaviours and optical performance were characterized and investigated in detail. Evidence shows that NaZnF3:Mn4+ exhibits a uniform particulate shape with single-phase crystallinity. By virtue of the non-equivalent substitution and the [MnF6] octahedral distortion in the fluoride host, sharp red emissions of phonon sidebands and the zero-phonon line upon blue light excitation are identified. Benefiting from the unique spectral feature, a wide colour gamut of 104.1% NTSC is achieved by coating β-SiAlON:Eu2+ and NaZnF3:Mn4+ on an InGaN chip, indicating the potential use of the Mn4+ fluoroperovskite as a colour converter for display backlight application.

Objective The present study investigated the potential correlation between non-high-density lipoprotein cholesterol/high-density lipoprotein cholesterol ratio (non-HDL-C/HDL) and the formation of coronary collateral circulation (CCC) in coronary artery disease cases with chronic total occlusive (CTO) lesions. Methods Two experienced cardiologists identified and selected patients with CTO lesions for retrospective analysis. The 353 patients were divided into a CCC poor formation group (Rentrop 0–1 grade, n = 209) and a CCC good formation group (Rentrop 2–3 grade, n = 144) based on the Cohen-Rentrop standard. A comparison of non-HDL-C/HDL ratios between the two groups was performed. The Spearman test was used to obtain the correlation between the cholesterol ratio and Rentrop grade. Independent predictors of CCC were analyzed using logistic regression. Receiver operating characteristic (ROC) curve analysis was also performed to quantify the predictive value of research indicator. Results The non-HDL-C/HDL ratio in the CCC poor formation group was elevated markedly compared to the CCC good formation group [( 3.86 ± 1.40) vs ( 3.31 ± 1.22), P = 0.000]. The Spearman test results indicated that non-HDL-C/HDL negatively correlated with Rentrop grade (r = − 0.115, P = 0.030). Multivariate logistic regression analysis showed that non-HDL-C/HDL ratio was an independent predictor of CCC formation (OR = 1.195, 95%CI = 1.020–1.400, P = 0.027). The area under the curve of ROC for detecting CCC poor formation was 0.611 (95% CI: 0.551–0.671, P = 0.000) with an optimal cut-off value of 2.77. Conclusion Non-HDL-C/HDL negatively correlated with the formation of CCC and served as an independent predictor of CCC formation, which may be used as a biomarker for the evaluation of CCC.

The photodetector is a key component in many applications, such as optical communication, image sensing, optical spectrometry, biological imaging and even radiation detection [ 1 – 6 ]. The silicon-on-insulator (SOI) technology has advantages, such as low leakage current, low parasitic capacitance and high resistance to radiation, and thus has been widely applied in integrated circuits with low power, high frequency and radiation-hardness [7 , 8] . Besides, SOI is widely used in photonics, thanks to the low loss of infrared light and the excellent ability to co-integrate electronic and photonic devices in the same chip. In this paper, we review several novel semiconductor photodetectors and image sensors developed by our group recently. These devices are fabricated on silicon-on-insulator (SOI) and two-dimensional (2D) substrates. They are based on novel device physics, and have extraordinary photodetection performances or unique functionalities compared to conventional devices.

In-memory computing may enable multiply-accumulate (MAC) operations, which are the primary calculations used in artificial intelligence (AI). Performing MAC operations with high capacity in a small area with high energy efficiency remains a challenge. In this work, we propose a circuit architecture that integrates monolayer MoS2 transistors in a two-transistor–one-capacitor (2T-1C) configuration. In this structure, the memory portion is similar to a 1T-1C Dynamic Random Access Memory (DRAM) so that theoretically the cycling endurance and erase/write speed inherit the merits of DRAM. Besides, the ultralow leakage current of the MoS2 transistor enables the storage of multi-level voltages on the capacitor with a long retention time. The electrical characteristics of a single MoS2 transistor also allow analog computation by multiplying the drain voltage by the stored voltage on the capacitor. The sum-of-product is then obtained by converging the currents from multiple 2T-1C units. Based on our experiment results, a neural network is ex-situ trained for image recognition with 90.3% accuracy. In the future, such 2T-1C units can potentially be integrated into three-dimensional (3D) circuits with dense logic and memory layers for low power in-situ training of neural networks in hardware., In standard computing architectures, memory and logic circuits are separated, a feature that slows matrix operations vital to deep learning algorithms. Here, the authors present an alternate in-memory architecture and demonstrate a feasible approach for analog matrix multiplication.

Background Endometrial cancer (EC) is the most prevalent gynecological malignant tumor worldwide. The prognosis of EC patients with distant metastasis is poor. Epithelial‐mesenchymal transition (EMT) plays a crucial role in tumor invasiveness and metastasis. the expression profile of EMT-related gene and their prognostic value and therapy target potential have not been systematically explored in endometrial cancer.MethodThe RNA-Seq expression profile and corresponding clinical data of EC patients were downloaded from The Cancer Genome Atlas database and the Gene Expression Omnibus (GEO) databases. A 9 EMT-related genes (EPHB2, TUFT1, CDKN2A, ONECUT2, RBP2, KLF8, E2F1, SIX1, ERBB2) prognosis risk model of EC was built by combining a machine learning algorism and multivariate Cox regression. Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) and Cibersort was applied to acquire for tumor purity and immune cell infiltration of high and low risk group. Also, response to immune checkpoint blockades (ICB) therapy and drug sensitivity were separately evaluated in TIDE and GDSC database for the EC patients.ResultA 9 prognostic EMT-related genes (ERGs) signature was constructed according to TCGA-UCEC and then verified in the GEO dataset. EC Patients were assigned to high- and low-risk group according to the signature, and significant survival difference were discovered between the two groups. Furthermore, ERGs signature was identified as an independent prognostic factor in the multivariate Cox regression and the ROC curve revealed a high sensitivity and specificity to the model (0.78 at 2 years and 0.83 at 5 years). Furthermore, quantitative prognostic nomogram was constructed to predict survival time in EC patients. This ERGs signature was also significantly correlated with tumor purity and immune cell infiltration, which in compliance to the prediction that clinical response to ICB therapy were better in the high-risk group base on TIDE database. In addition, 8 chemical components which the high-risk group may be more sensitive to were screened out and analyzed.Conclusions We identified a novel EMT‐related gene signature for predicting prognosis of EC patients and molecular understanding of EMT and immune microenvironment of EC. This may facilitate the decision-making during EC treatment.

Our study aimed to investigate the correlation of the imaging features obtained using conventional ultrasound (US) and elastography (conventional strain elastography of elasticity imaging [EI], virtual touch tissue imaging [VTI] and 2-D shear wave elastography [2-D-SWE] of virtual touch tissue imaging quantification [VTIQ]) with the clinicopathologic features and immunohistochemical (IHC) subtypes of breast cancer. The sample consisted of images from 202 patients with 206 breast lesions that were confirmed as breast cancers. Lesions with HER2 overexpression (luminal B HER2+ or HER2+) had higher mean shear wave velocity (SWV) values than the others. Older patients, lower histologic grade, no lymphovascular invasion and no lymph node metastasis were associated with luminal A (p < 0.001). There were significant differences in SWV values, histologic grade and lymph node status among the different pathologic types. This association may allow the use of 2-D-SWE in the pre-operative prediction of tumor characteristics and biologic activity, which may determine the prognosis in a non-invasive manner.

AIM: To explore a segmentation algorithm based on deep learning to achieve accurate diagnosis and treatment of patients with retinal fluid. METHODS: A two-dimensional (2D) fully convolutional network for retinal segmentation was employed. In order to solve the category imbalance in retinal optical coherence tomography (OCT) images, the network parameters and loss function based on the 2D fully convolutional network were modified. For this network, the correlations of corresponding positions among adjacent images in space are ignored. Thus, we proposed a three-dimensional (3D) fully convolutional network for segmentation in the retinal OCT images. RESULTS: The algorithm was evaluated according to segmentation accuracy, Kappa coefficient, and F(1) score. For the 3D fully convolutional network proposed in this paper, the overall segmentation accuracy rate is 99.56%, Kappa coefficient is 98.47%, and F(1) score of retinal fluid is 95.50%. CONCLUSION: The OCT image segmentation algorithm based on deep learning is primarily founded on the 2D convolutional network. The 3D network architecture proposed in this paper reduces the influence of category imbalance, realizes end-to-end segmentation of volume images, and achieves optimal segmentation results. The segmentation maps are practically the same as the manual annotations of doctors, and can provide doctors with more accurate diagnostic data.

To analyze the diagnostic value of adding SWE to MRI for the diagnosis of clinically significant prostate cancer with false-negative MRI results. This was a retrospective study of 367 patients who underwent MRI, SWE, and prostate biopsy between March 2016 and November 2018 at the Shanghai Tenth People’s Hospital. Serum prostate-specific antigen (PSA) and free PSA (fPSA) were measured preoperatively. Diagnostic value and accuracy was determined for MRI alone and MRI + SWE using the receiver operator characteristic curve (ROC) analysis. MRI misdiagnosed 17.9% (21/117) clinically significant prostate cancers, including 15 lesions in the peripheral zone and 6 in the central zone. Both qualitative and quantitative SWE could help detect 66.7% (10/15) significant prostate cancers with false-negative MRI, but there was no association with the Gleason score (p > 0.05). When considering the sextant of the peripheral zone, a significant association was not seen with histopathology in qualitative SWE (p = 0.071) and quantitative SWE (p = 0.598). Among age, PSA, fPSA, volume of the prostate gland, fPSA/PSA, and PSAD, only PSAD (p = 0.019) was associated with SWE results in patients with negative MRI. Adding SWE to MRI in patients with negative MRI for prostate examination could allow the correct diagnosis of additional patients and reduce the false-negative rate. • MRI plays an important role in clinically significant prostate cancers diagnosis. • SWE plays an important role in clinically significant prostate cancers with negative MRI. • Adding SWE to MRI in patients with negative MRI for prostate examination could allow the correct diagnosis of additional patients and reduce the false-negative rate.

In large-scale emotional events and complex emotional recognition applications, how to improve the recognition accuracy, computing efficiency, and user experience quality becomes the first problem to be solved. Aiming at the above problems, this paper proposes an emotional computing algorithm based on cross-modal fusion and edge network data incentive. In order to improve the efficiency of emotional data collection and the accuracy of emotional recognition, deep cross-modal fusion can capture the semantic deviation between multi-modal and design fusion methods through non-linear cross-layer mapping. In this paper, a deep fusion cross-modal data fusion method is designed. In order to improve the computational efficiency and user experience quality, a data incentive algorithm for edge network is designed in this paper, based on the overlapping delay gaps and incentive weights of large data collection and error detection. Finally, the edge network is mapped to a finite data set space from the set of emotional data elements inspired by heterogeneous emotional events. In this space, all emotional events and emotional data elements are balanced. In this paper, an emotional computing algorithm based on cross-modal data fusion is designed. The results of simulation experiments and theoretical analysis show that the proposed algorithm is superior to the edge network data incentive algorithm and the cross-modal data fusion algorithm in recognition accuracy, complex emotion recognition efficiency, and computation efficiency and delay.

OBJECTIVES The aim of this study was to evaluate the diagnostic performance of strain elastography, acoustic radiation force impulse (ARFI) imaging and point shear wave elastography (p-SWE) for assessment of the predominant types of intestinal stenosis in Crohn disease. METHODS Twenty-five patients were enrolled in this study, among whom 25 suspicious stenoses in 25 intestinal segments were studied using gray scale ultrasonography. All 3 elastography methods were performed, and all patients underwent endoscopy within 24 hours with pathologic biopsy. The sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), Youden index, and area under the receiver operating characteristic curve (AUROC) were calculated. Pathologic findings were regarded as the gold standard. RESULTS For SE, the optimal cutoff value was a score of 4 or greater (sensitivity, 75%; specificity, 66.7%; accuracy, 68%; PPV, 30%; NPV, 93.3%; AUROC, 0.708; however, P > .05). The results of ARFI imaging were similar: the optimal cutoff value was a score of 4 or greater (sensitivity, 50%; specificity, 81%; accuracy, 76%; PPV, 33.3%; NPV, 89.4%; AUROC, 0.669; P

To maintain fast development of communication toward terrahertz frequencies, InP-based high electron mobility transistors in InGaAs/InAlAs heterojunctions with ultra-short T-shape gates are constantly demanded. Our earlier work has successfully developed a 10 nm wide T-shape gate anchored by a Si3N4 sheath for high mechanical reliability, nicknamed as the tri-layer T-shape gate. However, the short channel effect, the gating efficiency and the sheath effect on the device performance based on this gate structure are still unclear. This work focuses on the above mentioned issues, trying to establish the connection between the gate structure and the device DC/AC performance through systematic simulation study. The simulation results show that the tri-layer T-shape gate not only enhances the mechanical reliability, but also reduces the lateral electric field around the gate-edges, leading to high break down voltage. Furthermore, such a gate structure exhibits an improved gating efficiency with a positive threshold voltage (Vth) for the gate-width down to 50 nm. The only drawback with the existence of the sheath is that the operation frequency is reduced by 17% comparing to the conventional T shape gate. However, as high as 550 GHz for the current cut-off frequency is still feasible, which is much higher than that with the Si3N4 passivation field plate gate. The foundation built in this work is extremely imperative to the advances of THz technology.

In this paper, the application of the zero subthreshold swing and zero impact ionization FET ( $\text{Z}^{2}$ -FET) for photodetection is studied with TCAD simulation. Dynamic coupling effect is utilized to form carrier injection barriers in the partially depleted silicon-on-insulator (PD-SOI) film. Photoelectron accumulation at the front gate interface lowers the hole injection barrier and modulates the turn-on voltage. The light-triggering threshold of the device can be tuned by the front gate voltage, which controls the injection barrier height. We explore two operation modes suited to different applications, and demonstrate the operation of a one-transistor active pixel sensor array. Unlike other image sensors that utilize only one type of carrier, the $\text{Z}^{2}$ -FET photodetector uses photo-generated holes to induce high electron currents through internal amplification, leading to a high sensitivity of up to $1.8\times 10^{5}$ e-/(lux $\cdot \text{s}$ ).

The benign development of Open-source Software Ecosystem (or OSSE) helps to fuse the wisdom of the community. It can facilitate the development and solve the urgent application needs of large-scale complex software systems. To guarantee that an OSSE is stable and effective for supporting the application development, health assessment for an OSSE has become a research hotspot. In this paper, starting from a new perspective, the OSSE is compared with the ecosystem in the natural world. An OSSE health measure method is proposed by integrating projection pursuit and real-coded accelerated genetic algorithm. First, according to the snowball sampling data collection method and the grounded theory, the data is collected and processed. Second, by designing evaluation indicators and utility functions, the projection pursuit classification model of the natural ecosystem is evaluated and combined with a real-coded accelerated genetic algorithm, thereby designing the health measure model. The experimental results suggest the effectiveness of the proposed approach.

Flexible lightweight supercapacitors are now actively investigated as one of the most promising power supplies for the portable and wearable electronics. Herein, radial nanowire arrays of Cu0.7Co2.3O4 are grown on the carbon fiber for a flexible electrode by hydrothermal process and subsequent thermal annealing. The Cu0.7Co2.3O4 electrode, with large exposure of low adsorption energy sites for Li ions due to the nanowires assembled by nanoparticles, displays superior electrochemical performance in 1 M LiCl aqueous solution and demonstrates areal capacitance of 340 mF cm−2 and 685Fg-1 at scan rate of 5 mV s−1. The contribution of Faraday process and capacitive process are analyzed in detail. The density functional theory is employed to calculate the Li+ ion adsorption energy and diffusion barrier, and the results indicate that Li ion can adsorb on 001 facet of the Co 2.3 Cu 0.7 O 4 easily and the Li ion diffusion barrier along 001 plane is as low as 0.16 eV. Furthermore, a flexible full solid-state supercapacitor is obtained by assembling two pieces of Co 2.3 Cu 0.7 O 4 electrode face to face, which presents good performance (112.5 mF cm−2 at 5 mV s−1), stability (74% after 5000 cycles), flexibility and wide operating voltage range. Three charged solid-state supercapacitors in series can light 48 green light emitting diodes for 150 s. The Cu0.7Co2.3O4 electrode is expected to be a potential candidate for actual applications.

In this paper, we demonstrate experimentally a photodetector based on silicon-on-insulator substrate. The photosensing diode is formed in the substrate induced by the backgate bias instead of doping. TCAD simulation is conducted to confirm that the field-induced junction performs similarly as the doped p-n junction in terms of photodetection. A simplified process flow with low process temperature is developed to fabricate the device, which can save implantation step and reduce the cost. The photoresponses of the device under various biases, light intensities, and wavelengths are studied systematically with both experiment and TCAD simulation. Responsivity up to 1150 A/W is achieved experimentally. In addition, the response spectrum can be modulated by the back gate bias, which is a very attractive feature for multiband photodetection application. The field-induced photodiode in the substrate modulates the threshold voltage of the MOSFET at top Si layer as efficiently as the doped junction, confirmed by TCAD simulation.

Summary Background & aims Epidemiological studies show inconsistent findings on the association of dietary fiber intake with risk of metabolic syndrome (MetS). Herein, we aim to conduct a meta-analysis of published studies to determine the role of dietary fiber in prevention of MetS. Methods A systematical search in PubMed and Embase databases through December 2016, together with reference scrutiny of relevant literature, was performed to identify studies for inclusion. We aggregated the odds ratios (ORs) with 95% confidence intervals (CIs) of MetS using a random effect model. Dose–response relationship between fiber intake and MetS was also evaluated. Results This meta-analysis included 8 cross-sectional and 3 cohort studies, totaling 28,241 participants and 9140 MetS cases. The highest versus lowest fiber intake was associated with a reduced risk of MetS (OR: 0.85, 95% CI: 0.79–0.92; P = 0.005), with moderate heterogeneity (I2 = 64%, P = 0.001) across studies. The benefit of fiber intake was significant among cross-sectional studies (OR: 0.85, 95% CI: 0.78–0.92; P Conclusions Dietary fiber intake is associated with less likelihood of having MetS. Additional large, prospective studies are warranted to enhance our findings.

This paper investigates the link selection policy for secure communications over a buffer-aided two-hop communication link. It is assumed that a source wishes to send information to a destination with the aid of a trusted half-duplex relay node compromised by an eavesdropper, and that there is no direct link between the source and the destination. The buffer-aided relay forwards the information to the destination by employing the randomize-and-forward scheme. Both the source and the relay transmissions are assumed to be vulnerable to the eavesdropper. The perfect channel side information of the network is assumed to be available at the source and the relay. Initially, conventional relaying protocols where equal partition of the time for the reception and transmission of the relay is considered. Then, the optimal time fraction with the optimal power control policy that maximizes the secrecy throughput is identified. Moreover, the optimal joint link selection and power control policy that maximizes the secrecy throughput is derived. Subsequently, a low complexity and asymptotically optimal link selection with ON/OFF power control is proposed. Numerical results demonstrate that the proposed link selection policies with power control can significantly improve the achievable secrecy throughput of the buffer-aided two-hop communication system.

A light-modulated subthreshold swing (SS) effect in a MoS2-Si hetero MOSFET is discovered. The device is essentially a SOI MOSFET with MOS2 top gate and partly gated channel. In the experiment, the SS of the device reduces from 100.2mV/dec to 70.4mV/dec under light illumination. In the TCAD simulation, the SS also shows similar light-modulated characteristic. The operation principle is mainly determined by the extra photovoltage at the top gate interface, as confirmed by the simulated results.

Using deep neural network algorithms, application scenarios in a distributed multi-data center environment, in response to fault location requirements, a distributed multi-center fault model is established, operation and maintenance data is analyzed, and key infrastructure states are modeled and located, carry out the digital representation of common failure scenarios to accelerate the positioning process of automatic operation.

Triggered by the pioneering research on graphene, the family of two-dimensional layered materials (2DLMs) has been investigated for more than a decade, and appealing functionalities have been demonstrated. However, there are still challenges inhibiting high-quality growth and circuit-level integration, and results from previous studies are still far from complying with industrial standards. Here, we overcome these challenges by utilizing machine-learning (ML) algorithms to evaluate key process parameters that impact the electrical characteristics of MoS2 top-gated field-effect transistors (FETs). The wafer-scale fabrication processes are then guided by ML combined with grid searching to co-optimize device performance, including mobility, threshold voltage and subthreshold swing. A 62-level SPICE modeling was implemented for MoS2 FETs and further used to construct functional digital, analog, and photodetection circuits. Finally, we present wafer-scale test FET arrays and a 4-bit full adder employing industry-standard design flows and processes. Taken together, these results experimentally validate the application potential of ML-assisted fabrication optimization for beyond-silicon electronic materials., Here, the authors demonstrate the application of machine learning to optimize the device fabrication process for wafer-scale 2D semiconductors, and eventually fabricate digital, analog, and optoelectrical circuits.

Phototreatment is at the leading edge of a research hot topic as a driving force for structural transformation, spectral and electromagnetism improvements, and the functional performance of nanomaterials. Light irradiation can excite surface plasmons in noble metal nanoparticles, create electron-hole pairs, and produce charge transfer in semiconductor substrates, which have led to it being widely used in surface-enhanced Raman spectroscopy (SERS) for life sciences, environmental protection, and biological analysis. Photo-induced enhanced Raman spectroscopy (PIERS) is a new technology developed on the basis of traditional SERS and has proven to be an efficient way to resolve several critical challenges thanks to its incomparable superiority for incontiguous operation, efficient charge separation and enrichment, and a large signal enhancement for a wide range of biomolecules at the trace level. This makes PIERS a powerful technique with very appealing and promising applications in various branches of analytical science. In this review, the enhancement mechanisms of PIERS are analyzed in comparison with SERS. Afterward, the parameters influencing the enhancement of PIERS, including the substrate, light irradiation, and relaxation are discussed in detail. Finally, some perspectives on further developments of PIERS are exemplified. The PIERS technique will continue to evolve and grow with new developments and its successful application in bioanalysis and life sciences.

Background Coronavirus disease 2019 (COVID-19) is an emerging infectious disease, first appeared in Wuhan, China, and quickly spread throughout the world. We aimed to understand the relationship between diabetes mellitus and the prognosis of COVID-19. Methods Demographic, clinical, laboratory, radiologic, treatments, complications, and clinical outcomes data were extracted from electronic medical records and compared between diabetes (n=84) and non-diabetes (n=500) groups. Kaplan-Meier method and multivariate Cox analysis were applied to determine the risk factors for the prognosis of COVID-19. Results Compared to non-diabetic patients, diabetic patients had higher levels of neutrophils (p = 0.014), c-reactive protein (p = 0.008), procalcitonin (p < 0.01), and D-dimer (p = 0.033), and lower levels of lymphocytes (p = 0.032) and albumin (p = 0.035). Furthermore, diabetic patients had a significant higher incidence of bilateral pneumonia (86.9%, p = 0.020). In terms of complications and clinical outcomes, the incidence of respiratory failure (36.9% vs. 24.2%, p = 0.022), acute cardiac injury (47.4% vs. 21.2%, p < 0.01) and death (20.2% vs. 8.0%, p = 0.001) in the diabetes group was significantly higher than that in non-diabetes group. Kaplan-Meier survival curve showed that COVID-19 patients with diabetes had a shorter overall survival time. Multivariate Cox analysis indicated that diabetes (HR 2.180, p = 0.031) was an independent risk factor for COVID-19 prognosis. In subgroup analysis, we divided diabetic patients into insulin required and non-insulin required groups according to whether they needed insulin, and found that diabetic patients requiring insulin may have a higher risk of disease progression and worse prognosis after the infection of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Conclusions Diabetes is an independent risk factor for the prognosis of COVID-19. More attention should be paid to the prevention and treatment for diabetic patients, especially those who require insulin therapy.

Medical named entity recognition (NER) tasks usually lack sufficient annotation data. Distant supervision is often used to alleviate this problem, which can quickly and automatically generate annotated training datasets through dictionaries. However, the current distantly supervised method suffers from noisy labeling due to limited coverage of the dictionary, which will cause a large number of unlabeled entities. We call this phenomenon an incomplete annotation problem. To tackle the incomplete annotation problem, we propose a novel distantly supervised method for Chinese medical NER. Specifically, we propose a high recall self-training mechanism to recall potential unlabeled entities in the distant supervision dataset. To reduce error in the high recall self-training, we propose a fine-grained lexicon enhanced scoring and ranking mechanism. Our method improves 3.2% and 5.03% compared to the baseline models on the dataset we proposed and a benchmark dataset for Chinese medical NER.

This is the system description of the CASIA_Unisound team for Task 1, Task 7b, and Task 8 of the sixth Social Media Mining for Health Applications (SMM4H) shared task in 2021. Targeting on deal with two shared challenges, the colloquial text and the imbalance annotation, among those tasks, we apply a customized pre-trained language model and propose various training strategies. Experimental results show the effectiveness of our system. Moreover, we got an F1-score of 0.87 in task 8, which is the highest among all participates.

Fructus Gleditsiae Abnormalis (FGA) has been used as a traditional Chinese medicine (TCM) for the treatment of stroke caused by phlegm and blood stasis. However, its substance basis and mechanism of action are currently unknown. This study is aimed to analyze the constituents of the volatile oil in FGA (VOFGA) using gas chromatography coupled with mass spectrometry (GC-MS) and explore the underlying effects and mechanisms of VOFGA in the prevention and treatment of ischemia stroke. An in vivo ischemia model was constructed by combination treatment of high-fat diet (HFD) and middle cerebral artery occlusion (MCAO) method. After administration, the cerebral infarction volume, the brain water content, hemorheology, blood lipids, inflammatory factors, oxidative stress indicators, Bax, Bcl-2, and cleaved caspase-3 and histological examination (HE) were determined and observed to explore the underlying effects and mechanisms of VOFGA against ischemia stroke. The results showed that forty components were determined after analyzed by GC-MS, and the percentage content of palmitate, paeonol, violetone, linalool, salpinol, citral, and methyleugenol were 4.69%, 5.2%, 3.56%, 3.31%, 2.42%, 2.65%, and 1.67%, respectively. The high dose of VOFGA could inhibit neurological damage; reduce the cerebral infarction volume and brain water content; improve whole blood viscosity and red blood cell aggregation index at various shear rates; reduce the levels of TG, TC, LDL-C, TNF-α, IL-1β, MDA, and NO; increase the contents of HDL-C, IL-10, and SOD; downregulate the expressions of Bax and cleaved caspase-3 in the ischemic regions; and upregulate the expressions of Bcl-2. These effects implied that VOFGA may exert neuroprotective effects via inhibiting ischemia-triggered oxidative damage-regulating blood lipid factors and reducing the production of proinflammatory mediators against cerebral I/R injury and neuronal apoptosis. The VOFGA presents a potential treatment value for cerebral ischemic stroke, and it may offer insights into discovering new active compounds for the treatment of ischemic stroke.

News recommendations aimed at alleviating network information overload. Traditional methods cannot simultaneously consider the representation of entities in the news at the level of knowledge and the influence of social networks on user interest points. The above two factors have a huge correlation with the efficiency of news recommendation. In order to consider these factors, this paper proposes a news recommendation model based on knowledge graph and social network which integrates knowledge graph representation and social networks into news recommendations. The model is a deep recommendation framework based on content and social networks for click-through rate prediction. The model utilizes the knowledge graph for representing entities in the news, quantifies the impact of social networks, and capture the dynamic changes of user interest. It adopts an improved sampling mechanism to quantify the social network structure. It uses a random walk sampling strategy to obtain neighbors in the social network. Moreover, it obtains the neighbor's influence weight on the target from interaction and content. The attention mechanism is used to quantify the effects of browsing records on user interests to capture dynamic changes. Experiments show that our model can effectively improve the effectiveness of news recommendations.