Search

Many breakthroughs on security and privacy-preserving techniques have emerged to mitigate the trust loss of cloud hosting environment caused by various types of attacks. To enhance memory-level security of multi-keyword fuzzy search, a widely occurred query request, we take the initiative to apply Trusted Execution Environment (a.k.a TEE) technology to our protocol design which provides hardware-based tamper-proof enclaves. Then we propose the Edit Distance-based Obfuscation Mechanism to further protect the query process executed outside TEE against access pattern leakage. With concerns of practicality and performance, we also propose the two-layer fuzzy index structure and Trend-aware Cache. The former addresses the space limitation of TEE memory for searching large datasets, while the latter optimizes the cache utility of TEE with trend-aware coordinator to effectively reduce the communication overhead.

Background Few studies have investigated psychogenic sexual dysfunction including psychogenic erectile dysfunction (pED); the effect of tadalafil on sexual behavior of male rats induced by chronic unpredictable mild stress (CUMS) remains unclear. Aim The aim was to explore the influence of CUMS on sexual behavior of male rats and the effects of tadalafil on that. Methods Adult male rats were divided into 3 groups, including the normal group without CUMS, the model group with 6 weeks’ CUMS, and the tadalafil group with treatment of tadalafil during CUMS. CUMS consists of water deprivation, food deprivation, stroboscopic lightning, white noise, cage tilting, weeding packing, and housing 2 unfamiliar rats. The apomorphine test and vaginal smear test were conducted with the aim to screen out male rats with good erectile function and make preparation for the sexual behavior test, respectively. Outcomes At the end of the study period, the level of anhedonia and sexual function were evaluated by the sucrose preference test, sexual behavior test, and measurement of serum testosterone, dopamine, and 5-HT. Results Sucrose preference showed significant decrease in rats after CUMS. The intromission ratio and total intromission frequency decreased significantly, while the mount latency and ejaculation latency prolonged significantly in CUMS-induced rats when compared with normal rats. Meanwhile, the treatment of tadalafil reversed the level of anhedonia and sexual function in CUMS-induced rats. However, there were no statistical differences in the levels of serum testosterone, dopamine, and 5-HT among groups. Clinical Implications The study constructed an animal model that can provide clinical insights into the mechanism of psychogenic sexual dysfunction and supports the application of tadalafil in pED therapy. Strengths and Limitations We found that CUMS-induced rats exhibited anhedonia and poor sexual function that could be prevented by tadalafil administration. Future research needs to construct the standard of pED model and explore the mechanism of tadalafil on central nervous system. Conclusion Tadalafil could prevent the changes of depression and poor sexual function in rats induced by CUMS, and the method of CUMS and the sexual behavior test should be used in the future for pED modeling.

By employing thiazole and 4-chlorothiazole as the A′ units, two A-D-A′-D-A type nonfused-ring electron acceptors (NFREAs) Tz-H and Tz-Cl were designed and synthesized. Replacing thiazole in Tz-H with 4-chlorothiazole can not only remarkably shorten the synthetic route through C-H direct arylation but also enhance molecular planarity with the simultaneous incorporation of S···N and S···Cl non-covalently conformational locks (NoCLs). The photovoltaic devices based on PM6:Tz-Cl exhibited a power conversion efficiency as high as 11.10%, much higher than that of PM6:Tz-H (6.41%), mainly due to more efficient exciton dissociation, better and more balanced carrier mobility, less charge recombination, and more favorable morphology. These findings demonstrate the great potential of NoCLs in achieving low-cost and high-performance NFREAs.

To make radial basis function (RBF) networks efficient for large-scale learning tasks, the parallel technique provides a promising way for the construction of multicolumn RBF network (MCRN), where the task to be tackled is decomposed into several subtasks and RBF networks designed for these subtasks are integrated as a parallel structure to obtain the final model. This paper focuses on presenting an efficient design method for MCRN based on improved generalized hybrid constructive (GHC) learning algorithm such that desired performance and high efficiency are guaranteed. Specifically, a minimum difference scheme is firstly introduced to divide a large-scale dataset into several parts with similar distribution. For each subtask, a compact RBF network is designed by improved GHC learning algorithm. By fully considering the inherent properties of RBF networks, a K-nearest potential centers based criterion is established to calculate the output of testing sample. Experiments on some benchmark classification problems are conducted to verify the advantages of the proposed method. In terms of training and testing efficiencies and classification accuracy, it is shown that our model is superior over some existing ones.

With the rapid development of deep learning, the performance on facial expression recognition (FER) has been greatly improved. However, multi-view FER, where the facial information is difficult to be fully acquired due to pose deflections, remains challenging. This paper presents a generative adversarial network (GAN) based three-stage-training algorithm for multi-view FER. Specifically, the frontal-view faces are firstly adopted to train a convolutional neural network (CNN) in an efficient way. Then, the pre-trained CNN is embedded in a well-designed GAN such that the faces with different deflections can be synthesized as the frontal-view ones containing expression features. The original and synthesized faces are finally fused and employed to retrain the CNN classifier such that the multi-view FER problem is efficiently resolved. Experimental results demonstrate that the three-stage-training algorithm can achieve better performance than some existing methods.

Background: Esophageal squamous cell carcinoma(ESCC) is a malignant tumor with high degree of malignancy worldwide, and the 5-year survival rate is only 20%. Previous evidence has linked the composition of gut bacteria to inflammation and the development of cancer. The purpose of this study was to investigate the relationship between intestinal flora composition and ESCC. In this study, fecal flora was used to represent gastrointestinal flora, and the specimens in the experimental group were stool of patients with ESCC confirmed by pathology. Stool samples from 13 patients with ESCC were collected as experimental group(ESCC), and stool samples from 13 healthy people were collected as control group(HC). There were 762 OTUs in the ESCC group and the HC group, 703 OTUs in the ESCC group and 418 OTUs in the HC group. Prevotella, butyricella and Satterella were the most abundant in the preoperative feces of patients with ESCC, while the most abundant in the feces of healthy people were cyanobacteria, pilonospiraceae and Sitterella. KEGG analysis showed that there were significant differences in membrane and intracellular structural molecules, cell antigens, toluene degradation, biosynthesis of ubiquitin and other terpanoid quinones, spindolipid biosynthesis and ganglia metabolic pathways between ESCC group and HC group. COG analysis showed that the ESCC group contained more peritinase, 2-succinyl-6-hydroxy-2, 4-cyclohexadiene-1-carboxylate synthetase, pyruvate: ferridoxin oxidoreductase and related 2-oxic acid: ferridoxin oxidoreductase β subunits. The occurrence and development of ESCC reduces the diversity of intestinal flora, and the composition of intestinal flora of ESCC is significantly different from that of HC, especially bacteria related to inflammation and lipid metabolism, which may become pathogenic bacteria promoting the development of ESCC.

Background Esophageal squamous cell carcinoma (ESCC) is the leading cause of cancer death in men and women worldwide. The poor prognosis and rapid increase in ESCC incidence highlight the need to promote early detection and prediction. Identifying key molecular targets involved in ESCC monitoring and progression is critical for ESCC patients. Methods This study examined miR-25/miR-203 as a biomarker for ESCC patients. Real-time quantitative polymerase chain reaction (PCR) was used to detect miR-25/miR-203 expression levels in tissues and serum exosomes, and MiR-25/miR-203 upregulation was confirmed in ESCC. Results We found that the miR-25/miR-203 ratio in cancer tissues from 36 ESCC patients was significantly enhanced compared with that in adjacent tissues. Moreover, the serum level of miR-25/miR-203 in 57 ESCC patients was higher than that in 31 healthy volunteers. Intriguingly, in 38 ESCC patients, the level of miR-25/miR-203 decreased significantly after surgery. Using ROC curve statistical analysis, we found that each group of miR-25/miR-203 had obvious sensitivity and high specificity. The miR-25/miR-203 relationship with the clinicopathological features of ESCC patients was also analyzed. MiR-25/miR-203 was significantly associated with the ESCC TNM-stage and lymph node metastasis, which predicts the prognosis of ESCC and reflects tumor progression. Conclusions This study highlights the feasibility of using exosome-mediated miR-25/miR-203 as a vital noninvasive biomarker for the detection and treatment monitoring of ESCC.

The growing number of mobile and IoT devices has nourished many intelligent applications. In order to produce high-quality machine learning models, they constantly access and collect rich personal data such as photos, browsing history, and text messages. However, direct access to personal data has raised increasing public concerns about privacy risks and security breaches. To address these concerns, there are two emerging solutions to privacy-preserving machine learning, namely local differential privacy and federated machine learning. The former is a distributed data collection strategy where each client perturbs data locally before submitting to the server, whereas the latter is a distributed machine learning strategy to train models on mobile devices locally and merge their output (e.g., parameter updates of a model) through a control protocol. In this article, we conduct a comparative study on the efficiency and privacy of both solutions. Our results show that in a standard population and domain setting, both can achieve an optimal misclassification rate lower than 20% and federated machine learning generally performs better at the cost of higher client CPU usage. Nonetheless, local differential privacy can benefit more from a larger client population ($>$> 1k). As for privacy guarantee, local differential privacy also has flexible control over the data leakage.

The thoracic duct is an important anatomical structure of the abdomen, chest and neck. An accurate understanding of the anatomy of the thoracic duct is critical to ensuring the safety and accuracy of the surgical procedure involving these areas, which can help surgeons reduce accidental injuries and provide more accurate diagnosis and intervention for patients with serious complications such as chylothorax, etc. In addition, it has great reference value to know the anatomy of thoracic duct in advance for the selection of treatment strategies in the treatment of refractory pleural effusion and ascites, chylothorax, and heart failure. Therefore, it is of practical value to be able to perform in vivo lymphangiography before surgery. However, the slender thoracic duct, the hidden position and the complicated anatomical structure variations pose a great challenge to the safety and accuracy of the surgical operation. This paper aims at a comprehensive discussion about anatomic variation of thoracic duct and the development of imaging techniques.

We propose a germanium (Ge) photonic crystal thin slab to achieve perfect absorption efficiency in telecom band. The maximum absorption is increased 14-fold compared to a planar cell with the same thickness at normal incidence. The absorption enhancement is attributed to the critical coupling with guided resonances by introducing an airhole photonic crystal on the thin slab. This novel design constitutes an important step towards high responsivity and high-speed photodetection with strong energy harvesting.

Near-infrared perfect wave harvesting of graphene is theoretically and numerically obtained in a hybrid dielectric configuration without assistance of a reflecting mirror. The absorption is increased 43-fold compared to a suspended graphene layer at normal incidence. The mechanism of perfect absorption is based on critical coupling with a guided resonance introduced by a silicon bar array and Fabry–Perot (FP) effect of a silicon oxide layer. This lossless design is expected to find applications to allow the active area with effective generation and fast transport of photocarriers, paving a new way for on-chip small-footprint ultrahigh responsivity and ultrahigh-speed photodetection in silicon photonics.

With the development of deep learning, the accuracy of face recognition of learning based approaches has even exceeded that of humans in some circumstances. However, identifying faces that are heavily deflected is still a challenging issue. To synthesize the frontal view from large-pose faces, a face merged generative adversarial network (FM-GAN) equipped with two generators and one discriminator is proposed in this paper. Generally, generative adversarial network (GAN) has only one generator and one discriminator to compete with each other to make the network convergence. By introducing an additional generator, the confrontation power is greatly enhanced and thus the performance of the designed model is improved. In the proposed framework, the first generator learns the upper and lower parts of a face to capture essential facial features. These high-dimensional information of the merged face together with the multi-scale encoded profile are transmitted into the second generator. Based on the competition with the discriminator, the final frontal face is synthesized by the second generator. To reduce the model complexity, FM-GAN only encodes the original image once through a pre-trained network, and the extracted features are shared by the two decoders. In addition, the first generator simultaneously produces the upper and lower parts of a face by the same encoder. Therefore, only the parameters of two decoders and one discriminator are required to be updated during the training process. The experimental results show that the frontal faces synthesized by our model can better preserve the facial identity and photorealistic than by some existing GANs.

BACKGROUND: To evaluate the safety, thoroughness and feasibility of “tunnel-type en bloc mediastinal lymph node dissection” in video-assisted thoracic surgery (VATS) for right non-small cell lung cancer (NSCLC) radical resection, which functionally dissected the lymph nodes of station 2R/4R/7. METHODS: A retrospective study was performed in the clinical data of 196 patients with VATS right NSCLC radical resection. According to the different methods of lymph node dissection of station 2R, 4R and 7, they were divided into the tunnel-type group (n=102) and the routine group (n=94). The clinical data of two group were compared. RESULTS: The analyses of the baselines of the two groups are comparable. For lymph nodes dissection of station 2R/4R/7, operation time, the total number, positive number and metastasis incidence shown no significant difference between two groups (P>0.05). However, the amount of bleeding, postoperative thoracic drainage volume, extubation time, hospitalization days, the incidence of postoperative pulmonary infection and chronic cough were significantly lower in the tunnel-type group (P

BACKGROUND This study aimed to investigate the safety and efficacy of minimally invasive total mesoesophageal excision (TME) for esophageal cancer. METHODS We retrospectively collected data from patients with esophageal cancer who underwent esophagectomy at our center between January 2011 and June 2017. Among 611 eligible patients, 302 underwent minimally invasive total mesoesophageal excision (the TME group) and 309 underwent non-total mesoesophageal excision (the NME group). Outcomes were compared after 1-to-1 propensity score matching, and subgroup analyses were performed for cases involving pT1-2 or pT3-4a disease. RESULTS The propensity score matching produced 249 pairs of patients. The TME group had a shorter operative time (P

In this paper, an efficient learning algorithm is proposed for complex-valued neural networks. It can well tune the weights and bias by complex-valued Levenberg-Marquardt (LM) algorithm and optimize the network structure by an incremental mechanism. Firstly, compared with the first-order algorithm, such as complex-valued gradient descent algorithm, complex-valued LM algorithm overcomes the problem of slow convergence for training complex-valued neural networks. Secondly, for practical applications, the choice of optimization algorithms is over-emphasized, while the effect of the network structure on performance is ignored. Indeed, the complexity of the structure is crucial to the generalization capability. Therefore, in order to improve the performance of complex-valued neural networks, an incremental mechanism is adopted to determine the network structure during the training process. Experimental results show the effectiveness of the proposed algorithm.

OBJECTIVE: The purpose of this paper was to investigate the role and mechanism of EEF1D in various diseases, especially in tumorigenesis and development, and explore the possibility of EEF1D as a biological target. BACKGROUND: EEF1D is a part of the EEF1 protein complex, which can produce four protein isoforms, of which three short isoforms are used as translation elongation factors. The three short isoforms play a role in anti-aging, regulating the cell cycle, and promoting the occurrence and development of malignant tumors, and the only long-form isoform plays a role in the development of the nervous system. METHODS: We searched the PubMed and Web of Science databases for literature up to January 2021 using relevant keywords, including “EEF1D”, “eukaryotic translation elongation factor 1 delta”, “translation elongation factor”, “translation elongation factor and cancer”, and “translation elongation factor and nervous system disease”. We then created an overview of the literature and summarized the results of the paper. CONCLUSIONS: Through the review of relevant articles, we found that EEF1D is obviously overexpressed in a variety of tumors, and can regulate the proliferation of tumor cells and tumor growth, as well as play a role in tumor invasion. EEF1D is likely to become a new biological target for tumor therapy and diagnosis.

Background Esophageal cancer (EC) is a malignant tumor with high mortality. Correlations have been found between the expression level of tropomyosin 3 (TPM3) and the depth of tumor invasion, lymph node metastasis, and the 5-year survival rate. However, the specific mechanisms underlying EC remain unclear. Methods Stably transfected TPM3-overexpresing and TPM3-knockdown esophageal squamous cell carcinoma (ESCC) cell lines (ECa109 and EC9706) were constructed, and the association between TPM3 and the proliferation, invasion, and migration of ESCC was investigated using molecular biology methods. The associations between TPM3 and matrix metalloproteinase (MMP)2/9 or epithelial-mesenchymal transition (EMT)-related proteins were verified, and the potential tumor-promoting mechanism was explored by Gelatin Zymography Experiment. Results TPM3 was found to promote the proliferation, migration, and metastatic potential of ESCC in vivo and in vitro, and stimulate the expression of MMP2/9 and certain EMT markers other than E-cadherin. The replenishment of MMP2/9 restored the malignant behavior of ESCC caused by TPM3. A gelatinase assay showed that the expression of TPM3 was related to the activity of MMP9. Conclusions TPM3 promoted proliferation, migration, and metastatic potential in EC cells. Additionally, TPM3 promoted the EMT process. This function may be achieved via the regulation the expression of MMP2/9.

Complex back-propagation (CBP) algorithm based on gradient descent is one of the most popular optimization algorithms for training complex-valued feedforward neural networks (CVFNNs). However, it has the disadvantages of slow convergence and easy to fall into local minimum. Most of previous investigations on CVFNNs focused on the improvement of optimization algorithms while the structure was determined by the trial-and-error method, which is a time-consuming process. To solve this issue, an efficient complex-valued hybrid incremental algorithm (CV-HIA) is proposed in this paper for three-layer split CVFNNs. The size of the underlying CVFNN is adaptively determined by an incremental scheme. When the training error falls into local minimum, hidden neurons are added into the network one by one. Nonlinear parameters between neurons of the hidden and input layers are adjusted by the gradient descent method. For linear weights between neurons of the output and hidden layers, the least-square (LS) method is adopted. Experimental results on several benchmark datasets show that the proposed method not only achieves good performance, but also obtains a compact network structure.

Oesophageal squamous cell carcinoma (ESCC) is an aggressive malignancy and a leading cause of cancer-related death worldwide. Lack of effective early diagnosis strategies and ensuing complications from tumour metastasis account for the majority of ESCC death. Thus, identification of key molecular targets involved in ESCC carcinogenesis and progression is crucial for ESCC prognosis. In this study, four pairs of ESCC tissues were used for mRNA sequencing to determine differentially expressed genes (DEGs). 347 genes were found to be upregulated whereas 255 genes downregulated. By screening DEGs plus bioinformatics analyses such as KEGG, PPI and IPA, we found that there were independent interactions between KRT family members. KRT17 upregulation was confirmed in ESCC and its relationship with clinicopathological features were analysed. KRT17 was significantly associated with ESCC histological grade, lymph node and distant metastasis, TNM stage and five-year survival rate. Upregulation of KRT17 promoted ESCC cell growth, migration, and lung metastasis. Mechanistically, we found that KRT17-promoted ESCC cell growth and migration was accompanied by activation of AKT signalling and induction of EMT. These findings suggested that KRT17 is significantly related to malignant progression and poor prognosis of ESCC patients, and it may serve as a new biological target for ESCC therapy. SIGNIFICANCE: Oesophageal cancer is one of the leading causes of cancer mortality worldwide and oesophageal squamous cell carcinoma (ESCC) is the major histological type of oesophageal cancer in Eastern Asia. However, the molecular basis for the development and progression of ESCC remains largely unknown. In this study, RNA sequencing was used to establish the whole-transcriptome profile in ESCC tissues versus the adjacent non-cancer tissues and the results were bioinformatically analysed to predict the roles of the identified differentially expressed genes. We found that upregulation of KRT17 was significantly associated with advanced clinical stage, lymph node and distant metastasis, TNM stage and poor clinical outcome. Keratin 17 (KRT17) upregulation in ESCC cells not only promoted cell proliferation but also increased invasion and metastasis accompanied with AKT activation and epithelial-mesenchymal transition (EMT). These data suggested that KRT17 played an important role in ESCC development and progression and may serve as a prognostic biomarker and therapeutic target in ESCC.

Background Interleukin‐1 promotes tumor angiogenesis through VEGF production. The interleukin‐1 receptor antagonist can suppress tumors by blocking this effect. Methods Immunohistochemistry, WB, and gene sequencing were used to analyze the expression of IL‐1RA in esophageal cancer patients. WB was used to detect the expression of IL‐1RA and interleukin‐1α in esophageal cancer cells. Stable ESCC cell models overexpressing the IL‐1RA were constructed. Their cell functions were tested, and their effects on VEGF were examined. Results IL‐1RA is downregulated in primary EC tumors, and this downregulation of IL‐1RA is closely related to TNM staging and survival prognosis. The overexpression of IL‐1RA increased the proliferation of KYSE410 EC cells, which have a high level of IL‐1α expression. Overexpression of IL‐1RA in KYSE410 cells promotes a decrease in the expression of VEGF‐A. However, IL‐1RA expression did not cause any changes in EC9706 cells with low IL‐1α expression. Conclusion IL‐1RA acts as a tumor suppressor, and its deletion promotes tumor progression by increasing VEGF‐A expression in ESCC.

Traditional Web data exchange research usually focuses on designing transformation rules but ignores the processing of the actual generated target data instances. Since the data instance is highly correlated with the schema and there are many duplicate elements in the source data instance, there is redundancy in the actual generated target data instance. In order to generate a target data instance solution that does not contain redundancy, under a given source-to-target exchange rule, a unified integration schema is designed firstly, and then, the instance block mechanism is introduced to analyze three mapping relationships of single homomorphism, full homomorphism and the isomorphism among the initial generated target data instances. According to the mapping relationship, three methods of instance selection, which are more compact, more informative and equivalence class processing, are proposed to remove redundant data instance in target data set and generate the core solution of target data instance. The experiment uses the data from the China Land Market Network to evaluate the performance of the data exchange core solution algorithm.

Existing algorithms in group recommendation usually focus on studying the rating data instead of the social tag data. Finding the group based on tags is a novel problem for group recommendation. In this paper, we first employ locality-sensitive hashing to quickly find potential user groups of similar types. Next, we adopt higher-order singular value decomposition to explore the potential semantic relevance among the ternary relationship data of and fill the sparse data of users for items. Finally, we apply a fairness aggregation strategy based on proportionality and envy-freeness to generate the fairest group recommendation list that is simulated to address the problem of finding the maximum coverage. We use the MovieLens dataset to demonstrate that the fairness strategy proposed here has a better fairness. What’s more, under the nDCG indicator, the algorithms based on the three strategies all generated a fairly good recommendation quality.

Skyline discovery from a hidden web database can enable a wide variety of innovative applications over one or multiple web databases in web information integration area. Due to the finite number of web accesses one can issue through per-IP-address or per-API-key, no more than k tuples of all matching tuples one top-k query can return, the restricted web interface types, etc., it is difficult to utilize traditional database Skyline computation methods for discovering Skyline of a hidden Web database. In the paper, we first analyze the key intersecting characters of skyline tuples by parallel coordinate system, and next define a search decomposition tree for searching skyline tuples of a hidden web database, and then prove that the tree is guaranteed to discover all skyline tuples. In the end, we put forward heuristic algorithms for getting skyline tuples of a hidden web database based on the mixture attributes. Theoretical analysis and real-world online and offline experiments demonstrate the effectiveness of our proposed methods and the superiority over baseline solutions.

Controversy about the adequate extent of lymph node (LN) dissection persists in surgery for thoracic esophageal squamous cell carcinoma (ESCC). The present study estimates the feasibility and strategy of common hepatic artery LN dissection during ESCC surgery. The clinical data of 482 patients with ESCC, who underwent thoracolaparoscopic esophagectomy at Fujian Medical University Union Hospital, were retrospectively selected. Among the 482 ESCC patients, 224 patients underwent thoracolaparoscopic esophagectomy with routine common hepatic artery LN dissection (cohort 1), while 258 patients underwent the same procedure without common hepatic artery LN dissection (cohort 2). The proposed operation method was introduced to safely dissect the common hepatic artery LN. Both univariate and multivariate analyses were performed to analyze the clinicopathological factors correlated to the common hepatic artery LN metastasis. The main postoperative complications were pneumonia, anastomotic leakage, vocal cord palsy and cardiovascular disease. There was no significant difference in the incidence of major postoperative complications between the 2 cohorts (P >.05), and the incidence was similar in a number of reports. The metastatic rate of common hepatic artery LNs was 4.91%, which was relatively lower. Based on the logistic regression analysis of 5 factors, tumor location and T classification were risk factors for common hepatic artery LN metastasis (P

This paper focuses on the creep characteristics of fly ash concrete corroded by sulfate. Firstly, based on the existing research, combined with the creep consolidation theory of concrete, the creep prediction model of fly ash concrete suitable for corrosive environment is analyzed. Secondly, through the analysis of the progress of sulfate corrosion of fly ash concrete, the concept of damage degree in damage theory is introduced to establish the creep calculation model of sulfate corrosion of fly ash concrete. Finally, the calculated value of the model is compared with the test value. The results show that the calculated value of the model is in good agreement with the experimental data and are consistent with the creeping growth trend.

One key challenge for data owners to host their databases in the cloud is data privacy. In this paper, we first demonstrate that even with the most recent hardware-based security technology such as Intel SGX, a hypervisor can still sniff key database operations running in its guest virtual machine (VM) such as the frequency and type of SQL queries, by monitoring the access pattern of this VM’s main and secondary memory. To ensure security against such access pattern monitoring attacks, we then propose ProDB, a minimal adaptation of a conventional DBMS with both hardware enclave and Oblivious RAM protocol. To enhance its performance for practical use, we also design a SQL-aware Path ORAM protocol called SaP ORAM, which optimizes the classic Path ORAM protocol under practical database workload. Through security analysis and extensive experimental results, we prove and show ProDB achieves high security and throughput on commodity cloud hosting servers.

Recurrent laryngeal nerve (RLN) injury is one of the most frequent postoperative complications of esophageal squamous cell carcinoma (ESCC) radical resection. This study aims to develop a novel scoring system to predict recurrent laryngeal nerve lymph node (RLNLN) metastases in early ESCC and explore the indications for precise RLN lymphadenectomy.Early stage ESCC patients from 2006 to 2014 were analyzed. Patient and pathologic characteristics were compared between patients with RLNLN metastases and those without. Univariate and multivariate analyses were performed to establish a scoring system that estimates the risks of RLNLN metastases. The indications for RLNLN dissection were validated by survival rate, postoperative complications, and metastases rate.A total of 311 cases selected from 1,466 ESCC patients were divided into the dissection group and the control group. Age, tumor length, macroscopic tumor type, T stage, tumor location and tumor differentiation were independent predictors of RLNLN metastases. The weighted scoring system included age (+2 for56 years), tumor length (+2 for over 4.45 cm), tumor location (+4 for upper thoracic, +2 for mid-thoracic) and macroscopic tumor type (+1 for advanced type). The total number of points estimated the probability of RLNLN metastases [low-risk (0-2 point), 0%; moderate-risk (3-4 points), 9.8%; and high-risk (4 points), 43.4%]. Besides, the dissection group had more complications and similar survival rate when compared with the control group.We developed a novel scoring system that accurately estimated the risk of RLNLN metastases in early ESCC patients. RLN lymphadenectomy may be safely omitted for the patients in the low-risk subgroup.

Convolutional neural networks (CNNs)have been proved to be excellent in image classification including facial expression recognition. However, CNNs commonly fail to learn the key expression characteristics with different facial angles. In this paper, the facial expression recognition problem is considered. The facial alignment technology is employed to adjust the inclined faces where an affine transformation is utilized such that the faces lie on a horizontal line. Then, in consideration of the scale of the model parameters, a sequential CNN is designed to extract the facial features, where a global averaged pooling layer is adopted instead of the fully connected layer. Comparing with the data augmentation, the experimental results on the three benchmark datasets (ie., Jaffe, CK+ and FER-2013)show that better performance is guaranteed by our model with less adjustable parameters.

A novel approach to investigate the assembly and fabrication method for Indium Arsenide nanowires nano devices. Assembly of InAs nanowires field-effect transistor (FET) was realized by dielectrophoresis approach. Before deposition of the contacts, InAs nanowires were treated wet etching in an ammonium polysulfide ((NH4)2Sx) solution to remove a surface oxide layer. The assembled devices were characterized by atomic force microscopy. The experimental results showed that the efficient assembly of InAs nanowires was obtained when the applied alternating current voltage has a frequency of 2MHz and an amplitude of 10V, and the success rate of ideal assembly for InAs nanowires FET had been assessed. Meanwhile, it also provided an effective assembly method for other one-dimensional nanomaterials assembly of nano devices.

With the continuous development of society, science and technology continue to enter this, China's education system is also in constant reform. From the reform of the situation, from the teaching idea to the evaluation system are all around the innovation education, innovation education as the center, and the teaching material evaluation system is the evaluation system is weak link, need to solve a lot of problems. This paper studies the construction of the evaluation system of the teaching material in innovative education, and analyzes the new mode and new method of the construction of innovative education system in Colleges and universities, in order to promote the development of innovation education in Colleges and universities. Keywords-Innovative Education; Teaching Material Evaluation System; Higher Education

In the final analysis, the competition among the modern society is the talent competition, the talented person is the importance of the country, and the analysis of the current international situation, it is necessary for our country to have a creative and innovative talents. The important task of innovation and entrepreneurship training is to fall in the major universities of our country, so it is very important to construct a new model. In the new mode, to cultivate students' innovation ability as the goal, through the cultivation and guidance of the individual, to open up a path for the students. In this paper, the main content of the training mode of innovative talents in Colleges and universities is to promote the cultivation of innovative talents in Colleges and universities in our country, and contribute to the cultivation of talents in China. Keywords-Individual Talents; Innovation and Entrepreneurship Talents; University Personnel Training Mode

It aims at enriching the gaming experience, setting the game free from the mouse and keyboard and converting the ordinary game into somatosensory game. Using motion capture technology, as the motion capture hardware monocular vision sensor, acquire the real-time movement of the moving object by the focus centroid algorithm to achieve the gesture recognition game controller. Experiments show that gesture recognition game controller designed in this paper is simple to operate, has real-time, high preparatory and takes up less system resources.