Search

Human action recognition is one of the hottest topics in the research field, so there are many relevant review papers illustrating the multi-modality of data, the selection of feature vectors, and the pros and cons of classification networks. With the continuous development of relational networks, graph convolutional networks (GCNs) have been applied to many different fields, including human action recognition. Although the graph convolutional networks have been demonstrated the powerful functionality in human action recognition, few literatures review GCN-based human action recognition. In this review, we not only give a detailed introduction to the structure of graph convolutional networks and data modalities used for human action recognition, but also focus on the application of GCNs in the field of human action recognition. Most importantly, we conduct experiments on five benchmark datasets for comparing the performance of seven state-of-the-art GCN-based algorithms for human skeleton-based action recognition. The five datasets selected in the experiments cover data of different scales (large-scale vs. small-scale) and different types (single-person, human-object interaction, and two-person-interaction) to explore the promising applicable scope of graph networks. We adopt the frequently used performance metrics such as accuracy, network parameters and loss function. Specifically, we analyze the impact of the multi-stream fusion strategies on improving the performance of the human action recognition schemes. To our best knowledge, it is the first time to survey human action recognition strategies related to GCNs, and to give a thorough experimental comparison of GCN-based human action recognition techniques with various datasets.

Generally, the anatomical CT/MRI modalities exhibit the brain tissue anatomy with a high spatial resolution, where PET/SPECT modalities show the metabolic features with low spatial resolution. Therefore, the integration of these two classes significantly improves several clinical applications and computer-aided diagnoses. In the proposed scheme, a fast local Laplacian filter (FLLF) is first applied to the source images to enhance the edge information and suppress the noise artifacts. Second, the RGB images are converted to YUV color space to separate the Y-component. Then to capture the spatial and spectral features of the source images, the NSST is applied to decompose the input (grayscale and/or Y-component) images into one low (LFS) and several high-frequency subbands (HFS). Third, an improved salience measure and matching factor (SMF) method by the local features-based fuzzy-weighted matrix (FW-SMF) is introduced to fuse the LFS coefficient. Due to the fast convergence with fewer iterations and robust pixels selection procedure, the PA-PCNN model is adopted to fuse the HFS coefficients. Fourth, the final fused image is obtained by applying inverse NSST and YUV format. Visual and statistical analysis performed on various experiments prove that the proposed scheme not only integrates the spatial and texture features details of the source images but also enhances the visual quality and contrast of the fused image compared to the existing state-of-arts.

We propose a novel single shot object detection network with a mask prediction branch. Our motivation is to enhance object detection features with semantic information extracted from deeper layers. The proposed mask prediction branch enriches important features in shallower layers with pixel-wise probability distribution of semantic information. Meanwhile, an improved receptive field block is adopted to increase the scale of receptive field of backbone network without too much extra computing burden. Our network improves the performance significantly over SSD and FSSD (Feature Fusion Single Shot Multi-box Detector) with just a little speed drop. In addition, we discuss the relationship between effective receptive fields and theoretical receptive fields on VGG16 backbone network. Comprehensive experimental results on PASCAL VOC 2007 demonstrate the effectiveness of the proposed method. We achieve a mAP of 79.8 with 300 × 300 input images (81.2 mAP by 512 × 512 inputs) at the speed of 58.4 FPS on a single Nvidia 1080Ti GPU. Experimental results demonstrate that the proposed network achieves a comparable performance with the state-of-the-arts.

The ECG signal is such a substantial means to reflect all the electrical activities of the cardiac system. Therefore, it is considered by the physician as the essential tools and materials to diagnose and treat heart diseases. To deal with different types of arrhythmia, the physician manually inspects the ECG heartbeat. Since there are tiny alternations in the amplitude, durations and therefore the morphology, the computer-based systems were needed to develop such solutions in order to help the physician to do their job. In this study, a novel tactic to automatically classify ten different arrhythmia types was developed depending on the deep learning theory. Consequently, the well-known convolutional neural network (CNN) approach was adopted to classify those different types of arrhythmia. The structure of the proposed model consists of 11 layers distributed as follows: four layers as convolution interchanged with other four layers of max pooling and finally three successfully connected layers. The experiment was conducted with the dataset which was downloaded from the Physionet in the Massachusetts Institute of Technology-Beth Israel Hospital database and then augmented to get sufficient and balanced dataset. To evaluate the performance of the proposed method and compare it with the previous algorithms, confusion matrix, sensitivity (SEN), specificity (SPE), precision (PRE), area under curve and receiver operating characteristic have been used and calculated. It has been found that performance from the proposed method is better than the existing methods based on CNN, and the accuracy is 99.84.

Elimination of combined Gaussian and impulse noises in digital image processing with preservation of image details and suppression of noise are challenging problem. For this purpose, a new filter which is median filters combined with convolutional neural network for Gaussian and salt & pepper noises. The previous methods are application dependents; some used for impulse noise and other employed only for Gaussian noise. The elimination of Gaussian and impulse noise completed into two steps. First the detection of impulse noise with the rejection of noise by employed of 3 × 3 and 5 × 5 window size median filters. In the second step removal of Gaussian noise performed by residual learning denoising convolutional neural network. It is very favorable and the ability of learning and denoising performance in the field of digital image processing. Denoising convolutional neural network also has active Gaussian noise with an unknown level of noise. Experimental work showed that the proposed method can achieve low loss and root mean square error during training, high peak signal to noise ratio, low mean square error, image quality assessment with good quality and mean absolute error for close prediction between denoised and original color images.

Smoke detection is of great significance for fire location and fire behavior analysis in a fire video surveillance system. Smoke image classification methods based on a deep convolution network have achieved high accuracy. However, the combustion of different types of fuel can produce smoke with different colors, such as black smoke, grey smoke, and white smoke. Additionally, the diffusion characteristic of smoke can lead to transparent smoke regions accompanied by colors and textures of background objects. Therefore, compared with smoke image classification, smoke region detection is a challenging task. This paper proposes a two-stream convolutional neural network based on spatio-temporal attention mechanism for smoke region segmentation (STCNNsmoke). The spatial stream extracts spatial features of foreground objects using the semi-supervised ranking model. The temporal stream uses optical flow characteristics to represent the dynamic characteristics of smoke such as diffusion and flutter features. Specifically, the spatio-temporal attention mechanism is presented to fuse the spatial and temporal characteristics of smoke and pay more attention to the moving regions with smoke colors and textures by predicting attention weights of channels. Furthermore, the spatio-temporal attention model improves the channel response of smoke-moving regions for the segmentation of complete smoke regions. The proposed method is evaluated and analyzed from multiple perspectives such as region detection accuracy and anti-interference. The experimental results showed that the proposed method significantly improved the ability of segmenting thin smoke and small smoke.

Background: Pancreatic cancer remains a highly fatal malignancy among the most common cancers. Multidisciplinary systemic treatment is the vital method to prolong the survival of patients with locally advanced and advanced disease. Case presentation: We report a patient presented to the hospital with upper abdominal pain with occasional diarrhea for 6 months, diagnosed as locally advanced pancreatic cancer (LAPC). The patient was detected carrying with BRAF p.N486_P490del mutation and positive EGFR expression, receiving induction sequential chemotherapy of gemcitabine plus nab-paclitaxel and modified fluorouracil plus leucovorin, oxaliplatin and irinotecan, combined with nimotuzumab, followed by consolidative carbon ion radiotherapy. After the sequential treatment, the efficacy evaluation was partial response and progression-free survival was at least 20 months. Conclusions: LAPC patients with non-V600 BRAF mutation and positive EGFR expression might benefit from induction sequential chemotherapy combined with nimotuzumab and consolidative carbon ion radiotherapy. It is possible for patient with locally advanced disease to obtain better local control and further survival.

ECG signals reflect all the electrical activities of the heart. Consequently, it plays a key role in the diagnosis of the cardiac disorder and arrhythmia detection. Based on tiny alterations in the amplitude, duration and morphology of the ECG, computer-aided diagnosis has become a recognized approach to classifying the heartbeats of different types of arrhythmia. In this study, a classification approach was developed based on the non-linearity and nonstationary decomposition methods due to the nature of the ECG signal. Complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) was used to obtain intrinsic mode functions (IMFs). Established on those IMFs, four parameters have been computed to construct the feature vector. Average power, coefficient of dispersion, sample entropy and singular values have been calculated as parameters from the first six IMFs. Then, ANN has been adopted to apply the feature vector using them and classify five different arrhythmia heartbeats downloaded from Physionet in the MIT–BIH database. To evaluate the performance of the proposed method and compare it with previous algorithms, confusion matrix, sensitivity (SEN), specificity (SPE), accuracy (ACC) and ROC have been used. It has been found that performance from the CEEMDAN and ANN is better than all existing methods, where the SEN is 99.7%, SPE is 99.9%, ACC is 99.9%, and ROC is 01.0%.

Background: Genetic factors play a crucial role in the glioma risk and prognosis of glioma patients. To explore the role of PVT1 polymorphism in the susceptibility and survival of glioma in the Chinese Han population, we conducted a case-control study.Methods: The three single-nucleotide polymorphisms (SNPs) in PVT1 were genotyped using Agena MassARRAY from 575 patients with glioma and 500 healthy controls. We used the χ 2 test to analyze the differences in distribution of allele and genotype between the cases and controls. Odds ratio (OR) and 95% confidence interval (95% CI) were calculated by logistic regression analysis to evaluate the association SNPs with glioma risk. The effects of polymorphisms and clinical features on survival of glioma patients were evaluated using the log-rank test, Kaplan-Meier and Cox regression analysis.Results: We found that rs13255292 was associated with a decreased risk of glioma in the recessive model in overall or male; and rs4410871 was significantly associated with an increased the risk of glioma in age≤40 years old or female. Moreover, the extent of resection and chemotherapy were found to be key prognostic factors in survival of glioma patients. However, the gender, age, tumor grade, radiotherapy and PVT1 polymorphisms have no effect on prognosis of glioma patients.Conclusions: Our results indicated that PVT1 polymorphisms (rs13255292 and rs4410871) were associated with glioma susceptibility, but have no effect on prognosis of glioma patients. Further studies with large samples are required to confirm the results.

Simple Summary Many wild felines are on the verge of extinction, and the monitoring of wildlife diversity is particularly important. Using surveillance videos of wild felines to monitor their behaviors has an auxiliary effect on the protection of wild felines. Through the actions of wild felines, such as standing, galloping, ambling, etc., their behaviors can be inferred and judged. Therefore, research on the action recognition of wild felines is of great significance to wildlife protection. The currently available methods are all aimed at experimental animals and design-specific feature descriptors for specific animals (such as color, texture, shape, edge, etc.), thus lacking flexibility and versatility. The proposed state-of-the-art algorithm using spatial-temporal networks combines skeleton features with outline features to automatically recognize the actions of wild felines. This model will be suitable for researchers of wild felines. Abstract Behavior analysis of wild felines has significance for the protection of a grassland ecological environment. Compared with human action recognition, fewer researchers have focused on feline behavior analysis. This paper proposes a novel two-stream architecture that incorporates spatial and temporal networks for wild feline action recognition. The spatial portion outlines the object region extracted by Mask region-based convolutional neural network (R-CNN) and builds a Tiny Visual Geometry Group (VGG) network for static action recognition. Compared with VGG16, the Tiny VGG network can reduce the number of network parameters and avoid overfitting. The temporal part presents a novel skeleton-based action recognition model based on the bending angle fluctuation amplitude of the knee joints in a video clip. Due to its temporal features, the model can effectively distinguish between different upright actions, such as standing, ambling, and galloping, particularly when the felines are occluded by objects such as plants, fallen trees, and so on. The experimental results showed that the proposed two-stream network model can effectively outline the wild feline targets in captured images and can significantly improve the performance of wild feline action recognition due to its spatial and temporal features.

Deep learning method is widely applied in modern artificial intelligence technology for Signal Modulation Recognition (SMR). Compared to CPUs and GPUs, FPGAs are highly energy-efficient and have low-latency streaming capabilities, which are more suitable for energy-sensitive or real-time machine learning projects. High-level synthesis (HLS) can automatically convert the logical structure described by a high-level language into a description by a low-level abstraction language. In this paper, we propose a system to optimize Deep Confidence Network (CDBN) by loops pipelining and unroll, memory buffering and partitioning, and implement an energy-efficient HLS-based FPGA Convolutional CDBN accelerator for SMR based on Virtex-7 platform. The accelerator system run at 150MHz and has 28% higher throughput and 80.5% less power consumption than a GPU implementation.

Bone mineral density (BMD) has been found to decrease in schizophrenia patients. We examined BMD and the levels of prolactin (PRL), bone alkaline phosphatase (BAP) and tartrate resistant acid phosphatase isoform 5b (TRACP-5b) in male chronic schizophrenia patients and compared them with healthy controls in a Chinese Han population, which has not been reported before. Male patients with chronic schizophrenia (SPs; n = 79) and healthy controls (HCs; n = 56) were recruited. BMD and plasma PRL, BAP and TRACP-5b levels were measured and compared between the two groups. The SPs group was further divided into two subgroups: the elevated PRL group (PRL ≥ 25 ng/ml, EPRL; n = 38) and the normal PRL group (PRL

Supplemental Digital Content is available in the text, The aim of this observational study was to test whether ABO blood type was a prognostic factor for pancreatic ductal adenocarcinoma (PDAC) patients and whether other risk factors could influence pancreatic cancer patients’ survival. This study included 610 patients who were diagnosed as pancreatic cancer and had undergone radical surgery. Patients’ characteristics included age, gender, tumor stage, tumor grade, adenosquamous carcinoma (ASC) status, preoperative serum carbohydrate antigen 19-9 (CA19-9) levels, preoperative serum carcinoembryonic antigen (CEA) levels, ABO blood type, smoking status, and drinking status were analyzed in this study. Cox proportional hazards regression model and Kaplan–Meier method were used to evaluate the role of prognostic factors. For pancreatic cancer patients undergoing radical surgery, the overall survival was worse for ASC patients than PDAC patients (Log-rank = 11.315, P

We aimed to evaluate the feasibility of interventional treatment of atrial septal defect (ASD) in low weight infants under 2-year-old. Seven hundred and ninety-three secundum ASD patients were divided into 2 groups: 665 were above 2-year-old and 128 were under 2-year-old. The basic conditions before the operation, postoperative complications within 24 hours, and adverse outcomes during a three-year follow-up were compared between the 2 groups using multivariate analysis. There were significant differences in age, weight, and the diameter of the ASD between the 2 groups (p0.001). The immediate success rate of the procedure was 96.7%. There were no significant differences in the success rate of the procedure, the incidence of residual shunt, arrhythmia, procedure-related arrhythmia, and occluder shedding between 2 groups (p0.05). Similarly, we found no association between age ≤2-year-old and any adverse outcomes postprocedure within 24 hours, including procedure failure (OR = 0.35; 95%CI: 0.04 to 2.93), residual shunt (OR = 1.07; 95%CI: 0.54 to 2.14), arrhythmia (OR = 0.68; 95%CI: 0.32 to 1.43), or procedure-related arrhythmia (OR = 0.34; 95%CI: 0.04 to 2.87). In the follow-up data, we found no association between age ≤2-year-old and arrhythmia (HR = 0.95; 95%CI: 0.50 to 1.80) and procedure-related arrhythmia (HR = 0.96;95%CI:0.25 to 3.64). Kaplan-Meier survival curves indicated no significant difference in the occurrence of arrhythmia between the 2 groups (log-rank test: p = 0.776). In conclusion, percutaneous ASD closure in young and low weight infants has a high success and low complication rate, along with reliable effects.

Digital image recognition has been used in the different aspects of life, mostly in object classification and detections. Monitoring of animal life with image recognition in natural habitats is essential for animal health and production. Currently, Sheep Pain Facial Expression Scale (SPFES) has become the focus of monitoring sheep from facial expression. In contrast, pain level estimation from facial expression is an efficient and reliable mark of animal life. However, the manual assessment is lack of accuracy, time-consuming, and monotonous. Hence, the recent advancement of deep learning in computer vision helps to classify facial expression as fast and accurate. In this paper, we proposed a sheep face dataset and framework that uses transfer learning with fine-tuning for automating the classification of normal (no pain) and abnormal (pain) sheep face images. Current state-of-the-art convolutional neural networks (CNN) based architectures are used to train the sheep face dataset. The data augmentation, L2 regularization, and fine-tuning has been used to prepare the models. The experimental results related to the sheep facial expression dataset achieved 100% training, 99.69% validation, and 100% testing accuracy using the VGG16 model. While employing other pre-trained models, we gained 93.10% to 98.4% accuracy. Thus, it shows that our proposed model is optimal for high-precision classification of normal and abnormal sheep faces and can check on a comprehensive dataset. It can also be used to assist other animal life with high accuracy, save time and expenses.

Background and purpose To evaluate the safety and efficacy of Total neoadjuvant treatment (TNT) in patients with rectal cancer with high risk factors. Methods and materials We did this phase 2 trial in patients who were diagnosed with stage II–III rectal cancer with at least one of the high risk factors. Three cycles of induction CAPOX were followed by pelvic radiotherapy of 50.4 Gy/28 fractions and two cycles of concurrent CAPOX. Three cycles of consolidation CAPOX were delivered after radiotherapy. Primary endpoints were pathological complete response (pCR) and R0 resection. Results Fifty patients were enrolled and 47 patients were evaluable. A total of 34 patients (72.3%) completed 6 to 8 cycles of chemotherapy and 46 patients (98%) completed the planned radiotherapy. 17 patients (36%) achieved a pCR or clinical complete response (cCR). Three cCR patients (6.4%) refused the operation and selected a watch-and-wait approach. The most common grade 3 or worse adverse events were leucopenia (10.6%) and radiation dermatitis (6.4%). The major surgical complications included pelvic abscesses/infection in 2 patients (4.3%), anastomotic leakage and hemorrhage in1 patient (2.2%), respectively, which were all addressed with conservative management. Conclusions TNT is effective and safe in patients with locally advanced rectal cancer with high risk factors. Long-term efficacies of TNT need to be further evaluated. This trial is registered with Chinese Clinical Trial Registry, number ChiCTR-OIN-17012284.

Background This meta-analysis of randomized controlled trials aimed to comprehensively assess the efficacy and toxicity of cyclin-dependent kinase (CDK) 4/6 inhibitors in advanced breast cancer (ABC) with hormone-receptor positive (HR+) and human epidermal growth factor receptor 2 negative (HER2−) disease. Methods We performed a systematical search using Cochrane Library, PubMed, Embase, and Web of Science up to March 2018. Only phase 2 and 3 randomized clinical trials assessing the efficacy and toxicity of the combination regimen of CDK4/6 inhibitors plus hormone therapy compared with hormone therapy alone were eligible for this meta-analysis. The pooled analyses of relative risk (RR) and hazard ratio were carried out by Stata software. Results A total of 7 randomized controlled trials including 3854 patients with HR+/HER2− ABC were included in this meta-analysis. The pooled hazard ratio for progression-free survival was 0.54 (95% confidence interval, 0.49-0.59; P Conclusion CDK4/6 inhibitors plus hormone therapy can significantly prolong the progression-free survival of patients with HR+/HER2− ABC and improve the objective response rate compared to conventional hormone therapy alone. The combined regimen results in a higher risk of AEs, especially grade 3/4 AEs.

Detecting camouflaged moving foreground objects has been known to be difficult due to the similarity between the foreground objects and the background. Conventional methods cannot distinguish the foreground from background due to the small differences between them and thus suffer from under-detection of the camouflaged foreground objects. In this paper, we present a fusion framework to address this problem in the wavelet domain. We first show that the small differences in the image domain can be highlighted in certain wavelet bands. Then the likelihood of each wavelet coefficient being foreground is estimated by formulating foreground and background models for each wavelet band. The proposed framework effectively aggregates the likelihoods from different wavelet bands based on the characteristics of the wavelet transform. Experimental results demonstrated that the proposed method significantly outperformed existing methods in detecting camouflaged foreground objects. Specifically, the average F-measure for the proposed algorithm was 0.87, compared to 0.71 to 0.8 for the other state-of-the-art methods., 13 pages, accepted by IEEE TIP

The purpose of image dehazing is the reduction of the image degradation caused by suspended particles for supporting high-level visual tasks. Besides the atmospheric scattering model, convolutional neural network (CNN) has been used for image dehazing. However, the existing image dehazing algorithms are limited in face of unevenly distributed haze and dense haze in real-world scenes. In this paper, we propose a novel end-to-end convolutional neural network called attention enhanced serial Unet++ dehazing network (AESUnet) for single image dehazing. We attempt to build a serial Unet++ structure that adopts a serial strategy of two pruned Unet++ blocks based on residual connection. Compared with the simple Encoder–Decoder structure, the serial Unet++ module can better use the features extracted by encoders and promote contextual information fusion in different resolutions. In addition, we take some improvement measures to the Unet++ module, such as pruning, introducing the convolutional module with ResNet structure, and a residual learning strategy. Thus, the serial Unet++ module can generate more realistic images with less color distortion. Furthermore, following the serial Unet++ blocks, an attention mechanism is introduced to pay different attention to haze regions with different concentrations by learning weights in the spatial domain and channel domain. Experiments are conducted on two representative datasets: the large-scale synthetic dataset RESIDE and the small-scale real-world datasets I-HAZY and O-HAZY. The experimental results show that the proposed dehazing network is not only comparable to state-of-the-art methods for the RESIDE synthetic datasets, but also surpasses them by a very large margin for the I-HAZY and O-HAZY real-world dataset.

The aim of the present meta-analysis was to systematically evaluate the overall diagnostic accuracy of F-FDG PET/CT for the detection of adrenal metastasis in lung cancer.A comprehensive literature search of Medline (PubMed), Embase, Web of Science, Cochrane, and China National Knowledge Infrastructure databases published through 10 January 2017 on the role of F-FDG PET/CT in evaluating adrenal mass in lung cancer patients was performed. Pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio of F-FDG PET/CT on a per lesion-based analysis were calculated. The area under the curve was calculated to measure the accuracy of F-FDG PET/CT.A total of nine studies, involving 707 lung cancer patients with 810 adrenal masses, were included in the present meta-analysis. Of 810 adrenal masses from lung cancer, 411 (50.7%) and 399 (49.3%) masses were benign and metastatic, respectively. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio of F-FDG PET/CT were 0.887 [95% confidence interval (CI): 0.852-0.917], 0.908 (95% CI: 0.875-0.934), 8.552 (95% CI: 6.197-11.802), 0.090 (95% CI: 0.039-0.211), and 96.825 (95% CI: 40.402-232.05), respectively. In addition, the area under the curve was 0.9622 and the overall diagnostic accuracy (Q* index) was 0.9077, suggesting excellent performance.F-FDG PET/CT plays a vital role in the differentiation of adrenal metastasis from a benign adrenal mass in lung cancer patients, with excellent diagnostic performance.

Arbitrary-order orbital angular momentum multiplexing (AOAMM) systems utilize OAM modes with both integer and fractional topological charges which are non-orthogonal. In this paper, the transmission matrix and the capacity per unit bandwidth, i.e., the spectral efficiency (SE) of an AOAMM system is derived based on the spatial cross correlations of the OAM submodes under different aperture conditions. The results show that in limited apertures, the SEs of AOAMM systems increase dramatically as the interval of two adjacent OAM submodes decreases by losing orthogonality. AOAMM systems are effective choices for satisfying the explosive growth of the communication requirements. This paper provides insight into the selection of spatially multiplexing approaches and the design of interference mitigation techniques for AOAMM systems with increased SEs.

Previous studies have demonstrated that neurotrophic factors may play a critical role in the severity of clinical symptoms in schizophrenia. However, it remains unknown whether serum levels of epidermal growth factor (EGF) in schizophrenia are similar to those observed in the case of other neurotrophic factors. Therefore, we compared serum EGF concentrations in first-episode drug-naive (FEP) patients and medicated chronic schizophrenic patients with healthy controls in order to explore whether EGF levels are related to psychopathological symptoms. We measured the serum levels of EGF in 78 first-episode medication-naive schizophrenia patients, 76 medicated chronic schizophrenic patients, and 75 healthy controls using the sandwich ELISA method. Disease severity were measured using the positive and negative syndrome scale (PANSS). Serum EGF levels showed a significant decrease in schizophrenia patients in comparison to healthy subjects. Serum EGF levels in FEP patients are indistinguishable from chronic cases. EGF levels were related to PANSS general symptom subscales in both FEP never-medicated and medicated patients. It is interesting that serum EGF levels were negatively correlated with the PANSS cognitive subscales, with the exception of the patients with chronic schizophrenia. Our preliminary results indicated that EGF may play a role in this illness and that it could be used as a potential biomarker of disease severity. Moreover, EGF may be associated with cognitive subscales of PANSS in FEP patients. Future studies should investigate the relationship between EGF and cognitive function as measured using standardized neuropsychological assessments to identify potential biomarkers related with cognition.

An improved algorithm for the fusion of anatomical structural MRI (magnetic resonance imaging) and metabolic functional PET (positron emission tomography) or SPECT (single photon emission computed tomography) images is proposed here. First, if any of the source image is in RGB format, it is converted to YIQ color space. Then by taking full advantage of non-subsampled shearlet transform (NSST), both the input images (channel-Y and/or gray-scale) are decomposed into one low frequency subbands (LFS) and several high frequency subbands (HFS). Second, local spatial frequency (SF) and region energy (RE) based fuzzy pixel-rules are applied for the fusion of LFS. Where, the fused HFS coefficient is directly selected by computing and comparing novel sum-modified-Laplacian (NSML) of every HFS, to extract maximum and more useful information. Finally, the required fused image is achieved by inversed NSST and inversed YIQ transform. Experimental results obtained shows that the algorithm significantly achieved maximum feature information from the source images, enhance visual quality and contrast of the fused image, and decrease computational task compared to other state-of-art proposed methods.

BackgroundAccumulating evidence suggests that serum vascular endothelial growth factor (VEGF) in many neurobiological processes potentially contributes to the pathophysiology of psychiatric disorders, particularly cognitive decline. The purpose of this study was to explore the differences in neurocognition, social cognition and VEGF among remitted first-episode schizophrenic patients, non-remitters and normal control subjects. Moreover, we investigated the association between serum VEGF levels and cognitive functions.Method65 remission (RS) and 45 nonremission patients (NRS) after first-episode schizophrenia, as well as 58 healthy controls (HC) were enrolled in this study. Social cognition was assessed using the Chinese Facial Emotion Test (CFET); neurocognition was measured with a test battery consisting of Hopkins Verbal Learning Test-Revised, Verbal Fluency Test, Trail Making Tests, Digit Span Tests (DST) and Stroop Tests. Blood samples were collected for VEGF measurements. Data was analyzed with SPSS 22.0 (Chicago, IL, USA).ResultsOn nearly all neurocognitive tests (except for DST), RS performed significantly worse than HC but better than NRS (P < 0.05). NRS, but not RS, exhibited markedly poorer social cognition than HC (except for Happiness and Surprise subscales of the CFET) (P RS > NRS).ConclusionCompared to HC, RS demonstrated poorer neurocognitive but intact social cognition functioning. These results indicate that VEGF levels decreased gradually with the severity of cognitive impairment in schizophrenia. VEGF may be involved in the pathological mechanism of cognitive performance in RS.

Background: Ki-67 is a widely used marker of tumor proliferation, but the prognostic value of ki-67 in triple-negative breast cancer (TNBC) has not been comprehensively reviewed. This meta-analysis was conducted to evaluate the association between ki-67 expression and survival of patients with resected TNBC.Materials and Methods: Relevant studies, evaluating the prognostic impact of pretreatment ki-67 in resected TNBC patients, were identified from PubMed, Embase, Web of Science, China National Knowledge Infrastructure, and Cochrane Library until March 14, 2019. Hazard ratios (HRs) with 95% confidence intervals (CI) were calculated as effect values for disease-free survival (DFS) and overall survival (OS).Results: In present meta-analysis, 35 studies with 7,716 enrolled patients were eligible for inclusion. Pooled results showed that a high ki-67 expression was significantly associated with poor DFS (HR = 1.73, 95% CI: 1.45–2.07, p < 0.001) and poor OS (HR = 1.65, 95% CI: 1.27–2.14, p < 0.001) in resected TNBC. In the subgroup analysis, when a cutoff of Ki-67 staining ≥40% was applied, the pooled HR for DFS and OS was 2.30 (95% CI 1.54–3.44, p < 0.001) and 2.95 (95% CI 1.67–5.19, p < 0.001), respectively.Conclusion: A high Ki-67 expression is a poor prognostic factor of resected TNBC. The cut-off of ki-67 ≥40% is associated with a greater risk of recurrence and death compared with lower expression rates, despite the Ki-67 threshold with the greatest prognostic significance is as yet unknown.

The aim of the present study was to evaluate the potential network of arsenic trioxide (ATO) target genes in pancreatic cancer. The DrugBank, STITCH, cBioPortal, Kaplan-Meier plotter and Oncomine websites were used to analyze the association of ATO and its target genes with pancreatic cancer. Initially, 19 ATO target genes were identified, along with their associated protein-protein interaction networks and Kyoto Encyclopedia of Genes and Genomes pathways. ATO was found to be associated with multiple types of cancer, and the most common solid cancer was pancreatic cancer. A total of 6 ATO target genes (namely AKT1, CCND1, CDKN2A, IKBKB, MAPK1 and MAPK3) were found to be associated with pancreatic cancer. Next, the mutation information of the 6 ATO target genes in pancreatic cancer was collected. A total of 20 ATO interacting genes were identified, which were mainly involved in hepatitis B, prostate cancer, pathways in cancer, glioma and chronic myeloid leukemia. Finally, the genes CCND1 and MAPK1 were detected to be prognostic factors in patients with pancreatic cancer. In conclusion, bioinformatics analysis may help elucidate the molecular mechanisms underlying the involvement of ATO in pancreatic cancer, enabling more effective treatment of this disease.

In this paper, the physical layer security (PLS) of fractional orbital angular momentum (OAM) multiplexing under atmospheric turbulence channels is studied. Based on the PLS theory, the secrecy capacities and the probabilities of positive secrecy capacities of fractional OAM (FrOAM) multiplexing systems with different topological charge intervals are analyzed. The influence of the eavesdropping ratio and the power allocation on secrecy capacities are compared. The simulation results show that, under the finite aperture limitation, the FrOAM multiplexing technique provides higher security over the integer OAM multiplexing in terms of the total secrecy capacities under weak and medium turbulence.

Background and purpose Manual delineation of clinical target volumes (CTVs) and organs at risk (OARs) is time-consuming, and automatic contouring tools lack clinical validation. We aimed to construct and validate the use of convolutional neural networks (CNNs) to set better contouring standards for rectal cancer radiotherapy. Materials and methods We retrospectively collected and evaluated computed tomography (CT) scans of 199 rectal cancer patients treated at our hospital from February 2018 to April 2019. Two CNNs—DeepLabv3+ for extracting high-level semantic information and ResUNet for extracting low-level visual features—were used for the CTV and small intestine contouring, and bladder and femoral head contouring, respectively. Contouring quality was compared using the paired t test. Five-point objective grading was performed independently by two experienced radiation oncologists and verified by a third. The CNN manual correction time was recorded. Results CTVs calculated using DeepLabv3+ (CTVDeepLabv3+) had significant quantitative parameter advantages over CTVResUNet (volumetric Dice coefficient, 0.88 vs 0.87, P = 0.0005; surface Dice coefficient, 0.79 vs 0.78, P = 0.008). Among 315 graded cases, DeepLabv3+ obtained the highest scores with 284 cases, consistent with the objective criteria, whereas CTVResUNet had the minimum mean manual correction time (7.29 min). DeepLabv3+ performed better than ResUNet for small intestine contouring and ResUNet performed better for bladder and femoral head contouring. The manual correction time for OARs was Conclusion CNNs at various feature resolution levels well delineate rectal cancer CTVs and OARs, displaying high quality and requiring shorter computation and manual correction time.

Background Transcatheter aortic valve implantation (TAVI) is an option for treatment for patients with severe aortic stenosis who are at high risk for death with surgical aortic valve replacement (SAVR). It is unknown whether TAVI can be safely introduced to intermediate- and low-risk patients. Objective To compare the efficacy and safety of TAVI and SAVR in patients with intermediate- and low-surgical risk. Data sources Medline, Embase, and the Cochrane Central Register of Controlled Trials were searched from inception to April 15, 2019. Study selection We included randomized controlled trials comparing TAVI with SAVR in patients with intermediate- and low-surgical risk. Data extraction Meta-analyses were conducted using random-effects models to calculate risk ratios (RR) with corresponding 95% confidence interval (CI). Two independent reviewers completed citation screening, data abstraction, and risk assessment. Primary outcome was a composite of all-cause mortality or disabling stroke at 12 months. Data synthesis A total of 5 trials randomizing 6390 patients were included. In patients with low risk, TAVI was associated with a significant reduction in the composite of all-cause mortality or disabling stroke compared with SAVR (RR, 0.56; 95%CI, 0.40-0.79; I2 = 0%). This benefit was not replicated in patients with intermediate risk (RR, 0.96; 95% CI, 0.80-1.15; I2 = 0%). Similar results were seen separately in all-cause mortality and disabling stroke when TAVI was compared with SAVR. Conclusion For patients with severe aortic stenosis who were at low risk for death from surgery, TAVI achieved superior clinical outcomes compared to SAVR; however, these benefits were not seen in those with intermediate risk. This information may inform discussions about deciding between SAVR and TAVI for patients with low to intermediate risk separately.

Rationale: Hepatocellular carcinoma (HCC), one of the most common cancers worldwide, is an aggressive tumor with very poor prognosis. Regorafenib was the first agent to show a survival benefit over placebo in patients who showed progression while being treated with sorafenib, but it remains an unsatisfactory agent owing to its serious side effects. Therefore, more efficient and milder therapies are needed. Patient concerns: Herein, we report a patient with advanced HCC with many lung metastases who showed progression during sorafenib treatment. Diagnoses: HCC with lung metastases (stage IVB). Interventions: SHR-1210 alone was used as second-line treatment. Outcomes: Although the lung metastases did not decrease 3 months after the treatment, they decreased significantly at 6 months after the treatment and partially disappeared. The tumor response indicated partial response. Furthermore, all of the lung metastases continued to decrease at about 17 months after treatment. The alpha-fetoprotein levels showed a similar trend. After a follow up of 19 months, the patient remains in good health. Lessons: SHR-1210 alone as a second-line treatment for a patient with HCC showed excellent antitumor effects. We think that SHR-1210 may exert its antitumor effects through a late-onset model, which persist for a long time. The side effects were mild and well tolerated.

e16040 Background: Surufatinib is a novel small-molecule kinase inhibitor targeting VEGFRs, FGFR and CSF-1R, simultaneous targeting of angiogenesis through VEGFRs/FGFR1 and modulating tumor immune microenvironment through CSF-1R may be a uniquely potent strategy to enhance antitumor activity. Toripalimab is a monoclonal humanized IgG4 PD-1 antibody. Encouraging efficacy of surufatinib plus toripalimab treating patients with advanced solid tumors was reported at 2020 AACR. This is an ongoing, multicenter, open-label, single-arm, phase II study to evaluate the efficacy and safety of surufatinib in combination with toripalimab in various solid tumors. Here we report the results of patients with advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma. Methods: Patients with histologically confirmed gastric or GEJ adenocarcinoma who have failed first-line of systemic chemotherapy were enrolled. Surufatinib 250 mg once a day (QD) will be orally administrated and toripalimab 240 mg will be intravenously administered every 3 weeks. Primary endpoint was the objective response rate (ORR) per RECIST v1.1. Results: As of Dec 31, 2020, a total of 21 gastric or GEJ adenocarcinoma patients were enrolled. The median age was 58 years old, and 81% of the patients were male. Median duration of treatment was 3 months (surufatinib, 3 months; toripalimab, 3 months). Among 15 patients with at least one post-baseline efficacy evaluation, 2 patients achieved confirmed partial response (PR), with 3 additional unconfirmed PR. And there were 6 in stable disease (SD), 3 in progressive disease (PD) and one not evaluable per RECIST v1.1. There were 5 in PR, 7 in SD and 2 in PD per irRECIST, respectively. The confirmed and unconfirmed ORR were 13.3% (95% CI: 1.7%-40.5%) and 33.3% (95% CI: 11.8%-61.6%), respectively. The disease control rate (DCR) was 73.3% (95% CI: 44.9%-92.2%) per RECIST v1.1. Median PFS was 3.71 months (95% CI: 1.41-unknown). 14.3% (3/21) of patients had treatment-related adverse events (TRAEs) of ≥ Grade 3. The most common TRAEs of ≥ Grade 3 were herpes zoster (4.8%), lymphopenia (4.8%), lymphocyte count decreased (4.8%), white blood cell count decreased (4.8%), liver injury (4.8%) and anaemia (4.8%). 4.8% (1/21) of patients had serious TRAEs. One patient died due to unknown reasons. Conclusions: Surufatinib plus toripalimab appeared to show encouraging activity in advanced gastric or GEJ adenocarcinoma with manageable safety profile. Such combination could be a promising strategy for advanced gastric or GEJ adenocarcinoma in the future. Clinical trial information: NCT04169672.

Purpose To evaluate the impacts of different leaf-width of multileaf collimator (MLC) on the fixed-beam intensity-modulated radiotherapy (IMRT) planning for prostate cancer patients with intermediate risk factors. Materials and methods Sixteen prostate cancer patients with intermediate risk factors were retrospectively recruited. For each patient, the two IMRT plans using MLC leaf width of 10 mm and 4 mm (sIMRT and mIMRT) were compared. All of plans were generated with the same setup and optimized parameters and methods. Homogeneity index (HI), conformity index (CI), dose distribution for planning tumor volume (PTV) and organs at risk (OARs) were analyzed to evaluate the quality of the plans. Delivery time (DT), monitor units (MUs) and control points (CPs) were used to assess the efficiency of treatment. Results For the target dose distribution, the mean HI values of the two plans for 16 patients are similar (sIMRT vs. mIMRT: 1.120±0.13 vs. 1.103±0.099, p>0.05). The mean CI of the mIMRT plans was significantly better than that of the sIMRT plans (sIMRT vs. mIMRT: 0.648±0.11 vs. 0.688±0.10, p Conclusions For the IMRT of the prostate cancer patients with intermediate risk factors, the plans with the narrower leaf-width MLC failed to promote the quality of plans, OARs dose sparing and the efficiency of treatment, and could only improve the conformity of the plans.

To obtain maximum information and key features from the source images, enhance visual quality and contrast of the fused image, and decrease computational task; an improved algorithm based on local featured fuzzy sets and NSML in NSST domain is presented here. First, by taking full advantages of NSST, two registered images of the same scene are decomposed into one Low Frequency Subbands (LFS) and several High Frequency Subbands (HFS). Then, Fuzzy Pixel-based fusion rules are applied on the LFS for computing weights of every pixel in the required fused coefficient. Weights are totally based on the local energies and entropies of the LFS. Where, fused HFS coefficients are selected by computing and comparing NSML of every HFS, to extract maximum and more useful information. Finally, inversed NSST is applied to get the required fused image. Additionally, the scheme is extended to color medical image fusion that effectively restrain color distortion and enhance visual quality. To assess the performance, several experiments were conducted on different data-sets of gray-scale and color medical images. Results obtained shows that the proposed algorithm is not only superior in edge and contour detection, visual feature and in computational performance but also presents an improvement in quantitative parameters compared to other state-of-art proposed schemes.

To investigate the treatment effects of gemcitabine plus S-1 (GS) for metastatic pancreatic cancer in our institution. Data from 41 patients with metastatic pancreatic cancer treated with GS regimen in West China Hospital, Sichuan University were reviewed. The therapeutic efficacy and toxicity were evaluated. The influencing factors of progression-free survival (PFS) and overall survival (OS) were also explored. At the last follow-up, all patients had died. The objective response rate was 22.0% (9/41) and the disease control rate was 65.9% (27/41). The median PFS and OS times were 5.1 (range, 1.5–21) and 10.6 months (range, 1.5–40), respectively. The 0.5-, 1-, and 2-year OS rates were 65.9%, 41.5%, and 9.8%, respectively. In multivariate analysis, body mass index and carbohydrate antigen 19-9 change were the significant influencing factors of PFS, compared to tumor site and chemotherapy cycles for OS. The adverse effects were moderate and tolerable. The effects of GS for metastatic pancreatic cancer in our institution were good. The adverse effects were moderate and tolerable. However, further investigation in future prospective clinical studies is warranted.

The most serious problems of OFDM are the high Out-Of-Band (OOB) Radiation and high Peak-to-Average Power Ratio (PAPR). So that it can’t use the fragmented spectrum and will also increase the system operating costs, reduce efficiency as well. Thus the Generalized Frequency Division Multiplexing (GFDM) comes into being as a candidate for the fifth generation (5G) wireless communication. GFDM can reduce the out-of-band radiation effectively but still suffer from pretty high PAPR. To solve this problem, studies have shown Clipping method can be used to reduce PAPR. However the computational complexity will increase along with the increase of the subcarrier number, and lower clipping threshold will cause severer impact on the signal distortion which is because the Clipping causes irreversible loss on large signals. In this paper, we propose a new method called majorizing compressing and expanding technique (MCT) to suppress the high PAPR of GFDM system. This method compresses high power signals, and expands low power signals at the transmitter and makes the inverse transformation at the receiver, so that the signal amplitude fluctuation is smaller and closer to the average, thus reduces the PAPR. Our simulations also reveal that there is a tradeoff between PAPR reduction and bit error rate(BER) performance. And a comparison between Clipping and MCT will be given to show that MCT performs better.

Smoke is an important indicator of forest fire video. However, many distractors such as heavy fog and smoke-like moving objects greatly degrade the recognition accuracy. This paper presents a novel smoke detection method of early forest fire video using CS Adaboost algorithm. First, motion regions are extracted from two adjacent frames by a proper background model which can avoid false positives of some static distractors, such as blue sky and gray leaves. Then, a CS Adaboost algorithm is used to recognize smoke regions using centroid movement by means of smoke flutter, image energy on the basis of the Wavelet Transform coefficients and color information between a reference smoke color and the input frame. Finally, the experimental results show that the proposed method can not only detect smoke image of early low thickness, but also more effectively distinguish dense fog from smoke.

Propose a flame region detection method using sparse representation of flame block.Establish a flame recognition model of based on mathematical model of meaning.Identify flame video from low-level visual features to high-level semantics.Validate higher accuracy of the proposed method with comparable computational time. The importance of flame detection cannot be ignored in a wildfire video surveillance system due to disturbance of heavy fog and challenging of smoke detection. In this paper a novel method for hierarchical detection of wildfire flame video is presented. Specifically, wildfire flame images are gradually recognized from low level visual features of pixel based to high level semantics of video clip based. For all the pixels of one image, the pixels which meet color rules and motion characteristics are labeled as flame colored pixels. The candidate flame region roughly generated by flame-like pixels is divided into non-overlapped image blocks. The sparse representation of the blocks are defined and recognized by learned dictionaries to more accurately segment candidate flame region and exclude some non-flame regions. To reduce the cost of computation, the proposed method detects one F rate ( F rate denotes one frame rate) frames instead of one frame at a time by using a sliding time window. Flicker features and spatiotemporal features extracted from video clips of the size F rate are used to build semantic model of wildfire flame video recognition based on mathematical model of meaning. Experimental results show that the proposed approach can effectively segment flame region and significantly improve the performance of wildfire flame detection.

An effective and prompt features extraction method plays a significant role in a forest fire detection system. However, exhaustively detecting all images in a long monitoring video is a serious time-consuming and simply extracting moving feature between two adjacent frames will greatly degrade the recognition accuracy. In this paper, a novel flame feature extraction method based on temporal-spatial domain is presented. A suspected flame video detected by color feature and moving feature is divided into a series of video clips based on the sliding time window. Color covariance descriptors and intensity descriptors are defined for representing color feature and texture feature of forest flame video. The width variation descriptor and the height variation descriptor are also defined for distinguishing flame videos from interfering videos such as red flowers video and red leaves video. Finally, the feature vector of a suspected flame video is extracted by computing co-covariance matrix composed of the above descriptors. The experimental results on flame detection demonstrate the effectiveness and usefulness of the proposed scheme.

In this paper, a new non-coherent chaos-based communication scheme, named orthogonal frequency division multiplexing based different chaos shift keying (OFDM-DCSK), is presented inspired by multi-carrier direct sequence code- division multiple access (MC-DS-CDMA) system, multi-carrier DCSK (MC-DCSK) system, and channel estimation of OFDM system. In this scheme, all of the occupied subcarriers are grouped into several groups. In each group, one chaotic reference is transmitted over the central subcarrier, while multiple modulated data streams are transmitted over the remaining subcarriers. Therefore, this transmitter structure increases the spectral efficiency, compared with the conventional DCSK system. At the receiver side, no radio frequency delay circuits are needed to demodulate the received data, which makes the system easy to implement in practice. The bit error rate (BER) performance of the OFDM-DCSK system is derived based on Gaussian approximation under additive white Gaussian noise (AWGN) channel. Simulation results are given under AWGN and two-path Rayleigh flat fading channel, which confirm the effectiveness of this new design.

Background Radiation pneumonitis (RP) is a common side reaction in radiotherapy for esophageal cancer. There are few reports about RP in esophageal cancer patients receiving postoperative intensity modulated radiation therapy (IMRT) and volumetric modulated arc therapy (VMAT). This study aims to analyze clinical or dosimetric factors associated with RP, and provides data for radiotherapy planning. Methods We reviewed 68 postoperative esophageal cancer patients who were treated with radiotherapy at the West China Hospital from October 2010 to November 2012 to identify any correlation between the clinical or dosimetric parameters and acute radiation pneumonitis (ARP) or severe acute radiation pneumonitis (SARP) by t-test, chi-square test, and logistic regression analysis. Results Of the 68 patients, 33 patients (48.5%) developed ARP, 13 of which (19.1%) developed SARP. Of these 33 patients, 8 (11.8%), 12 (17.6%), 11 (16.2%), and 2 (2.9%) patients were grade 1, 2, 3, and 4 ARP, respectively. Univariate analysis showed that lung infection during radiotherapy, use of VMAT, mean lung dose (MLD), and dosimetric parameters (e.g. V20, V30) are significantly correlated with RP. Multivariate analysis found that lung infection during radiotherapy, MLD ≥ 12 Gy, and V30 ≥ 13% are significantly correlated with an increased risk of RP. Conclusion Lung infection during radiotherapy and low radiation dose volume distribution were predictive factors associated with RP and should be accounted for during radiation planning.

Specific emitter identification (SEI) is a technique for distinguishing different emitters of a same type with other weak individual characteristics. Only using some traditional modulation parameters for recognition cannot distinguish different emitters with close modulation parameters. To solve the problem, new complex and high-dimensional features, which can characterize the emitters with more details, urgently need to be developed for recognition. An SEI method using fractal features based on box-counting dimension and variance dimension is presented. This paper mainly focuses on the weak individual characteristics caused by phase noise, applies fractal theory to the feature extraction, and finally establishes the recognition process using support vector machine. Numerical results show that the identification rate is generally more than 95% above 15dB of signal to noise ratio (SNR), and the real data experiment proves the practical performance of the proposed algorithm.

The ECG signals reflect all the electrical activities of the heart and are subjected to different kinds of noise. Hence denoising these signals is quite an important issue since the noise can hinder accurate diagnosis and detection of cardiac diseases. In this study, a novel strategy is proposed, based on Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) to remove Power Line Interference (PLI), with the performance being analyzed as well. The ECG signal is showed to be nonstationary and nonlinear. Therefore it is firstly decomposed by CEEMDAN to Intrinsic Mode Functions (IMFs), in order for problems of non-linearity and Mode Mixing issues in EMD must be solved. Then the first IMF is applied as a reference to different implementations of adaptive filters. Thus, solving the non-stationarity. Furthermore, by summing up the residual of the IMFs components, the first IMF is removed and the denoised signal constructed. The performance is compared with recent methods based on EEMD and measured by the Signal to Noise Ratio (SNR), correlation coefficient (CCR) and the Mean Square Error (MSE). Clean ECG records downloaded from the MIT-BIH Arrhythmia Database and the PLI noise are added and then denoised. It was founded that the SNR increased from 14.95241 to 27.295215 and the CCR increased from 0.984384 to 0.9991233 while MSE decreased from 0.141386 to 0.0341404, respectively for PLI noise. Results show that the proposed algorithm performs better than a recently developed algorithm.

Foreground detection has been widely studied for decades due to its importance in many practical applications. Most of the existing methods assume foreground and background show visually distinct characteristics and thus the foreground can be detected once a good background model is obtained. However, there are many situations where this is not the case. Of particular interest in video surveillance is the camouflage case. For example, an active attacker camouflages by intentionally wearing clothes that are visually similar to the background. In such cases, even given a decent background model, it is not trivial to detect foreground objects. This paper proposes a texture guided weighted voting (TGWV) method which can efficiently detect foreground objects in camouflaged scenes. The proposed method employs the stationary wavelet transform to decompose the image into frequency bands. We show that the small and hardly noticeable differences between foreground and background in the image domain can be effectively captured in certain wavelet frequency bands. To make the final foreground decision, a weighted voting scheme is developed based on intensity and texture of all the wavelet bands with weights carefully designed. Experimental results demonstrate that the proposed method achieves superior performance compared to the current state-of-the-art results., Comment: IEEE International Conference on Image Processing, 2017

Due to mass paper images and large amounts of noise in a paper image for a on-line paper detection system, this paper presents a novel method of on-line paper defect detection based on Multirepresentatives Classification (MRC). First of all, using the background subtraction method is used to rapidly identify those papers with defects from mass on-line papers. Afterwards, pixels of paper defect images are clustered to segmentalize defect regions, and LOG operator is used for edge extraction. On the basis of these, characteristic value of paper defect are extracted. Finally 8 kinds of paper defects are classified by using multi-representatives classification, the classification complexity of which is O(n). The experimental results showed that the method could quickly and accurately classify 8 kinds of paper defects, thus the method can meet the requirement of on-line paper defect detection.

BackgroundIntensive follow-up after surgery for colorectal cancers is common in clinical practice, but evidence of a survival benefit is limited.ObjectiveTo conduct a systematic review and meta-analysis on the effects of follow-up strategies for nonmetastatic colorectal cancer.Data sourcesWe searched Medline, Embase, and CENTRAL databases through May 30, 2018.Study selectionWe included randomized clinical trials evaluating intensive follow-up versus less follow-up in patients with nonmetastatic colorectal cancer.InterventionsIntensive follow-up.Main outcomes measuresOverall survival.ResultsThe analyses included 17 trials with a total of 8039 patients. Compared with less follow-up, intensive follow-up significantly improved overall survival in patients with nonmetastatic colorectal cancer after radical surgery (HR 0.85, 95% CI 0.74-0.97, P = 0.01; I2 = 30%; high quality). Subgroup analyses showed that differences between intensive-frequency and intensive-test follow-up (P = 0.04) and between short interval and long interval of follow-up (P = 0.02) in favor of the former one.LimitationsClinical heterogeneity of interventions.ConclusionsFor patients with nonmetastatic colorectal cancer after curative resection, intensive follow-up strategy was associated with an improvement in overall survival compared with less follow-up strategy. Intensive-frequency follow-up strategy was associated with a greater reduction in mortality compared with intensive-test follow-up strategy.

This study deals with the problem of reliable precise tracking control for the networked control system with non-zero external inputs (NEIs). A new control scheme is developed by introducing an integral item in the control law and taking stochastic actuator failure (SAF) into account, by which a better steady tracking performances can be achieved. Based on Lyapunov theory, a reliable control design method that guarantees the network-based tracking control systems with zero steady-state error in the conditions of both NEIs and SAFs is designed. Simulation results demonstrate the effectiveness of the developed controller design scheme.

M-ary variable period modulation is proposed in this paper as a candidate modulation scheme for an indoor visible light communication system. The principle of M-ary variable period modulation is described in details, the parameter selection guide and flicker mitigation and dimming control choices are discussed. Theoretical analysis and experimental results indicate that the method proposed is effective for an indoor visible light communication system.

Due to massive information and a high density of noise included in on-line web inspection detection system, a on-line paper defect Inspection method based on visual attention model is presented in this paper. Firstly, low-level of image feature at different scales are analyzed in the frequency domain. Then some corresponding salient maps are constructed, and combined into a salient image with normalization and linear-fusion in the spatial domain. Secondly, one viewpoints is found and used as the seed point for region growing in the image. After region growing, a paper defect region is extracted from background according to some rules. At last geometrical features and gray features of defect region are extracted, and imported into neural network for classifying paper defects. The experimental results showed that the method had high identification rate, and performed well even in on-line paper images with a high density of noise.

In this study, granular sorbent (GS) which was prepared from red mud, clay and coal by sintering method was studied. With the aim of determining the optimal parameters for GS production, the influences of several important factors (including raw material ratio, sintering temperature and sintering time) on the adsorption capacity of GS for Cu2+ removal were evaluated in the single-experiment. The results indicated that the raw material ratio was the key factor that influenced the property of GS. The optimal conditions for GS preparation were as follows: the raw material ratio (red mud: clay: coal) of 85:10:5, sintering temperature of 400 °C and sintering time of 20 min.