Search

How to locate the target function faster and more accurately is a key problem of Automatic Reverse-Engineering of Software Programs. In order to solve this problem, a target function location method based on code coverage analysis is proposed. Firstly, it obtains function call information of the program and calculate the suspicious rate of each function. And then a stability factor is proposed to reduce the noise functions. Finally, the target function is successfully located. The experimental results show the method proposed has linear time complexity. For software programs with million function calls, it can accurately locate the target function within several minutes, the performance and accuracy are greately improved compared to the contrast methods.

In order to solve the problems that face recognition models based on deep learning are difficult to deploy in embedded devices and their real- time performance is poor, the existing model compression and acceleration algorithms are deeply studied, and a neural network compression algorithm based on knowledge distillation and adversarial learning is proposed. The algorithm framework consists of three parts: a large-scale teacher network for pre-training, a lightweight student network and a discriminator for adversarial learning. This paper improves the traditional knowledge distillation loss by adding an indicator function, so that the student network only learns the classification probability that the teacher network correctly identifies. Since the feature map of the middle layer has rich high-dimensional features, the discriminator in adversarial learning strategy is introduced to identify the difference between the student network and the teacher network in the feature map level. Furthermore, in order to improve the generalization ability of the student network and enable it to be applied to different machine vision tasks, the teacher network and the student network learn from each other in the latter part of the training and update alternately, so that the student network can explore its own optimal solution. The results are verified on CASIAWEBFACE and CelebA data sets respectively. The experimental results show that the recognition accuracy of the small size student network obtained by knowledge distillation is only about 1.5% lower than that of the teacher network with full supervision training. At the same time, the proposed method is compared with the feature map-oriented knowledge distillation algorithm and the model compression algorithm based on adversarial learning training, and the proposed method has a better performance. [ABSTRACT FROM AUTHOR]

Recent works based on deep learning and facial priors have performed well in superresolving severely degraded facial images. However, due to the limitation of illumination, pixels of the monitoring probe itself, focusing area, and human motion, the face image is usually blurred or even deformed. To address this problem, we properly propose Face Restoration Generative Adversarial Networks to improve the resolution and restore the details of the blurred face. They include the Head Pose Estimation Network, Postural Transformer Network, and Face Generative Adversarial Networks. In this paper, we employ the following: (i) Swish-B activation function that is used in Face Generative Adversarial Networks to accelerate the convergence speed of the cross-entropy cost function, (ii) a special prejudgment monitor that is added to improve the accuracy of the discriminant, and (iii) the modified Postural Transformer Network that is used with 3D face reconstruction network to correct faces at different expression pose angles. Our method improves the resolution of face image and performs well in image restoration. We demonstrate how our method can produce high-quality faces, and it is superior to the most advanced methods on the reconstruction task of blind faces for in-the-wild images; especially, our 8 × SR SSIM and PSNR are, respectively, 0.078 and 1.16 higher than FSRNet in AFLW. [ABSTRACT FROM AUTHOR]

In the process of face recognition, face acquisition data is seriously distorted. Many face images collected are blurred or even missing. Faced with so many problems, the traditional image inpainting was based on structure, while the current popular image inpainting method is based on deep convolutional neural network and generative adversarial nets. In this paper, we propose a 3D face image inpainting method based on generative adversarial nets. We identify two parallels of the vector to locate the planer positions. Compared with the previous, the edge information of the missing image is detected, and the edge fuzzy inpainting can achieve better visual match effect. We make the face recognition performance dramatically boost. [ABSTRACT FROM AUTHOR]

Random testing (RT) is widely applied in the area of software testing due to its advantages such as simplicity, unbiasedness, and easy implementation. Adaptive random testing (ART) enhances RT. It improves the effectiveness of RT by distributing test cases as evenly as possible. Fixed Size Candidate Set (FSCS) is one of the most well-known ART algorithms. Its high failure-detection effectiveness only shows at low failure rates in low-dimensional spaces. In order to solve this problem, the boundary effect of the test case distribution is analyzed, and the FSCS algorithm of a limited candidate set (LCS-FSCS) is proposed. By utilizing the information gathered from success test cases (no failure-causing test inputs), a tabu generation domain of candidate test case is produced. This tabu generation domain is eliminated from the current candidate test case generation domain. Finally, the number of test cases at the boundary is reduced by constraining the candidate test case generation domain. The boundary effect is effectively relieved, and the distribution of test cases is more even. The results of the simulation experiment show that the failure-detection effectiveness of LCS-FSCS is significantly improved in high-dimensional spaces. Meanwhile, the failure-detection effectiveness is also improved for high failure rates and the gap of failure-detection effectiveness between different failure rates is narrowed. The results of an experiment conducted on some real-life programs show that LCS-FSCS is less effective than FSCS only when the failure distribution is concentrated on the boundary. In general, the effectiveness of LCS-FSCS is higher than that of FSCS. [ABSTRACT FROM AUTHOR]

We aimed to value the diagnostic potential of serum miR-1297 in esophageal squamous cell cancer (ESCC). Its expression level was detected in 156 pairs of patients with ESCC and healthy volunteers using quantitative real-time polymerase chain reaction (qRT-PCR) method. It was statistically decreased in ESCC patients compared with healthy controls. AUC based on serum miR-1297 was 0.840 ± 0.035 in discovery group and 0.837 ± 0.034 in validation group. Further analysis on early-stage patients revealed that the AUC was 0.819 ± 0.053 in discovery group and 0.814 ± 0.044 in validation group. Its sensitivity and specificity were promising. In conclusion, serum miR-1297 can serve as an ideal indicator for the diagnosis of ESCC. [ABSTRACT FROM PUBLISHER]

Lung cancer is currently the leading cause of cancer-related death worldwide and it is important to identify the predictive and/or prognostic markers for the cancer. Nestin , a proliferative and multipotent biomarker has been reported to be associated with prognosis in non-small cell lung cancer (NSCLC) in a few studies. In the present study, we retrospectively recruited 153 patients with NSCLC. Nestin protein expression in tumor samples was determined by immunohistochemistry staining. Nestin expression was related with tumor differentiation (P=0.036), lymphatic metastasis (N stage, P=0.011), and p-TNM stage (P=0.013), while there was no significant association between Nestin expression level and age, smoking habits, gender, histologic type, and T stage. Nestin was an independent prognostic factor for overall survival in NSCLC with an adjusted hazard ratio of 2.701 (95% CI, 1.616-4.513, P<0.001) after controlling the confounding factors. Then we determined the effects of Nestin on cell proliferation, colony formation, invasion, and apoptosis by knockout of Nestin with a new developed method, CRISPR/Cas9 mediated genome editing. It was observed that knockout of Nestin caused enhancement of cancer cell apoptosis and inhibition of cell proliferation, colony formation, and invasion in A549 and H1299 cell lines. Furthermore, we examined the expression of epithelial-mesenchymal transition (EMT) related biomarkers such as E-cadherin and Vimentin in Nestin -depleted lung cancer cells and knockout of Nestin was found to inhibit EMT, suggesting the involvement of Nestin mediated EMT signaling in lung cancer. The finding above demonstrated that Nestin might serve as a prognostic factor and therapeutic target in NSCLCs.

We aimed to investigate the pattern of expression and clinical significance of isocitrate dehydrogenase 1(IDH1) in esophageal squamous cell carcinoma (ESCC). The IDH1 expression was determined by quantitative real-time polymerase chain reaction, immunohistochemistry, and Western blot analysis using 38 pairs of frozen tissues. Enzyme-linked immunosorbent assay was employed to measure 67 pairs of serum samples from patients and their controls to evaluate its diagnostic value. Immunohistochemistry analysis of 111 formalin-fixed paraffin embedded tissue samples was conducted for explaining its prognostic value. After shRNA transfection, CCK8 and clonal efficiency assays were carried on for verifying the function of IDH1 in vitro. Increased expression at mRNA (P < 0.001) and protein levels (immunohistochemistry: P < 0.001, Western blot analysis: P < 0.001) were observed. Similarly, the IDH1 expression in serum from patients with ESCC was significantly upregulated relative to that from healthy controls (P < 0.001). Kaplan-Meier curve indicated that IDH1 upregulation predicted worse overall survival (OS) and progression-free survival (PFS). Univariate and multivariate analyses identified IDH1 expression as an independent prognostic factor for OS and PFS. Furthermore, OD450 values and colony numbers were decreased in sh-IDH1 groups (all P < 0.05). In conclusion, IDH1 is upregulated in patients with ESCC and can be used as a good potential biomarker for diagnosis and prognosis.

Up to now, the prognosis of non-small cell lung cancer (NSCLC) is poor. With progress of cancer biology, a number of genes have been investigated for predicting prognosis of NSCLC, such as cancer stem cell markers SRY (sex determining region Y)-box 2 (SOX2) and Nestin. Recently, a series of studies have been performed to examine the associations of SOX2 and Nestin with clinical parameters and prognosis in NSCLC, however, the results were not consistent. In the present study, we conducted a systematic review and meta-analysis to summarize the associations. Four English databases (PubMed, ISI web of science, Embase, and Ovid) were used to search the relevant studies with the last date of November 10, 2015. The pooling analyses were stratified by DNA amplification and protein expression. The pooling ORs or HRs were used to assess the strength of the associations. Finally, we included 19 articles for SOX2 and six articles for Nestin according to the inclusion and exclusion criteria. The pooling analyses revealed that there were significant associations between SOX2 DNA amplification and clinical features of NSCLC, gender, smoking status, squamous cell cancer (SCC) histology, and differentiations. And significant associations were also identified between SOX2 protein expression and clinical parameters, smoking status and SCC histology. For Nestin, its protein expression was correlated with lymph node metastasis and stage. Simultaneously, we found that high/positive SOX2 alterations, either DNA amplification or protein expression, were favorable for overall survival (OS) in NSCLC. On the contrary, high/positive Nestin protein expression was poor for OS., Competing Interests: The authors declare no conflicts of interest.

Non‑small cell lung cancer (NSCLC) remains the leading cause of cancer‑related mortality despite the fact that great advances have been made in therapeutic treatment methods. Therefore, in the present study, the role of adenovirus-mediated retinoblastoma 94 (Ad‑RB94) gene therapy in NSCLC was investigated. Following treatment with Ad‑RB94, the proportion of A549 cells in the G2/M phase was increased. In the mouse xenograft model, the overexpression of RB94 inhibited the tumor growth compared with the control group and the Ad-‑LacZ-treated group. In the transplanted tumors, the overexpression of RB94 induced the apoptosis of tumors as well as an increase in the mRNA levels of cyclinB1. In conclusion, the results of the present study suggested that RB94 may effectively inhibit NSCLC tumor cell growth by inducing G2/M cell cycle arrest and apoptosis, indicating that RB94 may be a promising candidate for adjuvant therapy with radiation or chemotherapy in NSCLC.

Peroxisome proliferator-activated receptor β/δ (PPARβ/δ) is an essential transcription factor in myocardial metabolism. This study aims to investigate the effects of PPARβ/δ activation in the adult heart on mitochondrial biology and oxidative metabolism under normal and pressure-overload conditions. We have investigated the effects of cardiac constitutively active PPARβ/δ in adult mice using a tamoxifen-inducible transgenic approach with Cre-LoxP recombination. The expression of PPARβ/δ mRNA and protein in cardiomyocytes of adult mice was substantially increased after short-term induction. In these mice, the cardiac expression of key factors involved in mitochondrial biogenesis, such as PPARγ coactivator-1, endogenous antioxidants Cu/Zn superoxide dismutase, and catalase, fatty acid, and glucose metabolism, such as carnitine palmitoyltransferase Ib, carnitine palmitoyltransferase II, and glucose transporter 4, were upregulated. Subsequently, myocardial oxidative metabolism was elevated concomitant with an increased mitochondrial DNA copy number and an enhanced cardiac performance. Moreover, activation of PPARβ/δ in the adult heart improved cardiac function and resisted progression to pathological development in mechanical stress condition. We conclude that PPARβ/δ activation in the adult heart will promote cardiac performance along with transcriptional upregulation of mitochondrial biogenesis and defense, as well as oxidative metabolism at basal and pressure-overload conditions.