A Novel Anti-Flashover Superhydrophobic Coating with Self-Assembly Characteristic of Surface Energy Differences.

Because of the versatility of superhydrophobic materials, they have attracted a lot of attention even in power electronics, transportation, engineering, and other fields. The volume fraction of fluorinated silicon oxide nanoparticles in superhydrophobic materials is one of the most important factors. Increasing the volume fraction will decrease the stability between the coating and the hydrophobic surface. Especially, the flashover voltage of the coating gradually decreases from 10 to 35 vol.%. Meanwhile, the flashover voltage dispersion of the coating increases drastically after 30 vol.%. In order to improve the electrical properties of the superhydrophobic coating, self-assembly of surface energy differences strategy is proposed in this work. A binary filling phase of the coating is introduced by 2D boron nitride nanosheets and silicon oxide nanoparticles. Although Hexagonal boron nitride with high surface energy and low roughness, it will be spontaneously assembled and wrapped by silicon oxide nanoparticle based on surface energy differences, which forming a low surface energy filled phase. Experiment results prove that the flashover voltage of the superhydrophobic coating is optimized by the binary filling phase coating. This method offers new ideas for the selection of filling phase and application of superhydrophobic materials.

HYAL-1-induced autophagy facilitates pancreatic fistula for patients who underwent pancreaticoduodenectomy.

The main mechanism of hyaluronidase 1(HYAL-1) in the development of postoperative pancreatic fistula (POPF) after pancreatoduodenectomy (PD) was unknown. In this study, a comprehensive inventory of pre-, intra-, and postoperative clinical and biological data of two cohorts (62 pancreatic cancer [PCa] and 111 pancreatic ductal adenocarcinoma [PDAC]) which could induce POPF were retrospectively analyzed. Then, a total of 7644 genes correlated with HYAL-1 was predicted in PDAC tissues and the enriched pathway, kinase targets and biological process of those correlated genes were evaluated. Finally, a mouse pancreatic fistula (PF) model was first built and in vitro studies were performed to investigate the effects of HYAL-1 on PF progression. Our data indicated that preoperative serum HYAL-1 level, pancreatic fibrosis score, and pancreatic duct size were valuable factors for detecting POPF of Grade B and C. The serum HYAL-1 level of 2.07 mg/ml and pancreatic fibrosis score of 2.5 were proposed as the cutoff values for indicating POPF. The bioinformatic analysis and in vitro and in vivo studies demonstrated that HYAL-1 facilitates pancreatic acinar cell autophagy via the dephosphorylation of adenosine 5'-monophosphate-activated protein kinase (AMPK) and signal transducers and activators of transcription 3 (STAT3) signaling pathways, which exacerbate pancreatic secretion and inflammation. In summary, the preoperative serum HYAL-1 was a significant predictor for POPF in patients who underwent PD. Tumor-induced HYAL-1 is one of core risk in accelerating PF and then promoting pancreatic secretion and acute inflammation response through the AMPK and STAT3-induced autophagy.

Circular RNA hsa_circ_0005909 modulates osteosarcoma progression via the miR-936/HMGB1 axis.

BACKGROUND: Osteosarcoma (OS) is the most common bone malignant tumor in children, youth, and adolescents. Circular RNA hsa_circ_0005909 (circ_0005909) is involved in the progression of OS. Nevertheless, there are few reports on the role and mechanism of circ_0005909 in OS. METHODS: Quantitative real-time polymerase chain reaction (qRT-PCR) was executed to examine the expression of circ_0005909, miR-936, and High Mobility Group Box 1 (HMGB1) mRNA in OS tissues and cells. Cell viability, colony formation, migration, and invasion were evaluated by Cell Counting Kit-8 (CCK-8), cell colony formation, or transwell assays. Cell epithelial-mesenchymal transition (EMT) and HMGB1 protein levels were assessed through western blot analysis. The role of circ_0005909 on tumor growth in vivo was verified by xenograft assay. The relationship between circ_0005909 or HMGB1 and miR-936 was confirmed with the dual-luciferase reporter or RNA pull-down assays. RESULTS: Circ_0005909 level was upregulated in OS tissues and cells. OS patients with high circ_0005909 expression had a lower survival rate. Circ_0005909 inhibition reduced tumor growth in vivo and constrained cell viability, colony formation, migration, invasion, and EMT of OS cells in vitro. Furthermore, circ_0005909 served as a sponge for miR-936 and the repressive impacts of circ_0005909 silencing on malignant behaviors of OS cells were abolished by miR-936 inhibitors. Also, HMGB1 acted as a target for miR-936 and was modulated by circ_0005909 via miR-936. Additionally, HMGB1 overexpression restored the inhibitory influence on the malignant behaviors of OS cells mediated by circ_0005909 inhibition. CONCLUSIONS: Circ_0005909 inhibition impeded the progression of OS via downregulating HMGB1 via sponging miR-936.

Soluble urokinase plasminogen activator receptor associates with higher risk, advanced disease severity as well as inflammation, and might serve as a prognostic biomarker of severe acute pancreatitis.

BACKGROUND: This study aimed to explore the potential of soluble urokinase plasminogen activator receptor (suPAR) as a biomarker for severe acute pancreatitis (SAP) risk prediction and disease management in SAP patients. METHODS: Totally 225 acute pancreatitis (AP) patients (including 75 SAP, 75 moderate-severe acute pancreatitis [MSAP], and 75 mild acute pancreatitis [MAP] patients) were recruited based on the Atlanta classification, and their serum samples were obtained within 24 hours after admission. Meanwhile, 75 health controls (HCs) were recruited with their serum samples collected at the enrollment. The serum suPAR was then detected using enzyme-linked immunosorbent assay. RESULTS: The suPAR level was increased in SAP patients compared with MSAP patients (P = .023), MAP patients (P < .001), and HCs (P < .001). Receiver operating characteristic (ROC) curve presented that suPAR could not only differentiate SAP patients from HCs (AUC: 0.920, 95%CI: 0.875-0.965) but also differentiate SAP patients from MSAP (AUC: 0.684, 95%CI: 0.600-0.769) and MAP patients (AUC: 0.855, 95%CI: 0.797-0.912). In SAP patients, suPAR was positively correlated with Ranson score (P < .001), acute physiology and chronic healthcare evaluation II score (P = .001), sequential organ failure assessment score (P < .001), and C-reaction protein (P = .002). Further ROC curve exhibited that suPAR (AUC: 0.806, 95%CI: 0.663-0.949) was of good value in predicting increased inhospital mortality of SAP patients. CONCLUSION: Soluble urokinase plasminogen activator receptor is of good predictive value for SAP risk and may serve as a potential biomarker for disease severity, inflammation, and inhospital mortality in SAP patients.

Plasma D-Dimer Level Is an Early Predictor of Severity of Acute Pancreatitis Based on 2012 Atlanta Classification.

BACKGROUND Acute pancreatitis (AP) is a common digestive disorder. Its management depends on the severity; therefore, it is essential to stratify AP patients early. D-dimer, a coagulation indicator, appears to be associated with the pathogenesis of AP. The aim of this study was to evaluate D-dimer as an early predictor of the severity of AP. MATERIAL AND METHODS This was a single-center retrospective study of 1260 patients diagnosed based on the revised Atlanta classification. Only patients hospitalized within 24 h of onset were included, and 334 patients were enrolled. Blood was collected at admission and 3 times within 48 h of admission. Values at admission and average of the 3 blood samples were evaluated by univariate and multivariate analyses. Furthermore, the area under the receiver-operating characteristic curve (AUC) was used to estimate the validity of the predictor and to define optimal cut-off points for prediction. RESULTS We found that 53.3% of the patients had mild AP (MAP), 24.3% had moderately severe AP (MSAP), and 22.4% had severe AP (SAP). D-dimer at admission and the average D-dimer could distinguish MAP patients from MSAP and SAP patients, with cut-off values of 3.355 mg/L and 4.868 mg/L, respectively. No difference in the parameters at admission was observed in multivariate analysis in distinguishing SAP from MSAP, but the average D-dimer level was significantly different with a cut-off value of 7.268 mg/L by comparing Ranson score, APACHE II score, and D-dimer level. CONCLUSIONS The average value of D-dimer levels could be used as a predictor of severity of AP. In general, patients with an average D-dimer level <4.868 could be diagnosed with MAP, >7.268 would develop into SAP, and between 4.868 and 7.268 would be MSAP.

Crucial microRNAs and genes in metformin's anti-pancreatic cancer effect explored by microRNA-mRNA integrated analysis.

Despite great improvements in surgical procedures and chemotherapy, pancreatic cancer remains one of the most aggressive and fatal human malignancies, with a low 5-year survival rate. Therefore, novel therapeutic strategies for the prevention and treatment of pancreatic cancer are urgently needed. The present study aimed to investigate the mechanisms by which metformin exerts its anticancer effects on the microRNA-mRNA interactions in human pancreatic cancer. Microarray and systematic analyses revealed that the anti-pancreatic cancer effects of metformin were correlated with 3 up-regulated microRNAs and 4 of their target mRNAs. In addition, the microarray and systematic analyses ultimately demonstrated that 3 microRNAs regulated 4 key mRNAs in a sub-pathway of pancreatic cancer and then affected growth, angiogenesis, and apoptosis. This finding may provide a deeper understanding of the mechanisms by which metformin suppresses proliferation and angiogenesis and promotes apoptosis in pancreatic cancer cells. Collectively, this experiment improves the understanding of the mechanisms by which metformin suppresses pancreatic cancer and indicates that metformin, the most commonly used drug for the treatment of diabetes mellitus, may be a promising candidate agent for the treatment of pancreatic cancer.

Hydrogen sulphide exacerbates acute pancreatitis by over-activating autophagy via AMPK/mTOR pathway.

Previously, we have shown that hydrogen sulphide (H2 S) might be pro-inflammatory during acute pancreatitis (AP) through inhibiting apoptosis and subsequently favouring a predominance of necrosis over apoptosis. In this study, we sought to investigate the detrimental effects of H2 S during AP specifically with regard to its regulation on the impaired autophagy. The incubated levels of H2 S were artificially intervened by an administration of sodium hydrosulphide (NaHS) or DL-propargylglycine (PAG) after AP induction. Accumulation of autophagic vacuoles and pre-mature activation of trypsinogen within acini, which indicate the impairment of autophagy during AP, were both exacerbated by treatment with NaHS but attenuated by treatment with PAG. The regulation that H2 S exerted on the impaired autophagy during AP was further attributed to over-activation of autophagy rather than hampered autophagosome-lysosome fusion. To elucidate the molecular mechanism that underlies H2 S-mediated over-activation of autophagy during AP, we evaluated phosphorylations of AMP-activated protein kinase (AMPK), AKT and mammalian target of rapamycin (mTOR). Furthermore, Compound C (CC) was introduced to determine the involvement of mTOR signalling by evaluating phosphorylations of downstream effecters including p70 S6 kinase (P70S6k) and UNC-51-Like kinase 1 (ULK1). Our findings suggested that H2 S exacerbated taurocholate-induced AP by over-activating autophagy via activation of AMPK and subsequently, inhibition of mTOR. Thus, an active suppression of H2 S to restore over-activated autophagy might be a promising therapeutic approach against AP-related injuries.

Significantly reduced radiation dose to operators during percutaneous vertebroplasty using a new cement delivery device.

BACKGROUND: Percutaneous vertebroplasy (PVP) might lead to significant radiation exposure to patients, operators, and operating room personnel. Therefore, radiaton exposure is a concern. The aim of this study was to present a remote control cement delivery device and study whether it can reduce dose exposue to operators. METHODS: After meticulous preoperative preparation, a series of 40 osteoporosis patients were treated with unilateral approach PVP using the new cement delivery divice. We compared levels of fluoroscopic exposure to operator standing on different places during operation. group A: operator stood about 4 meters away from X-ray tube behind the lead sheet. group B: operator stood adjacent to patient as using conventional manual cement delivery device. RESULTS: During whole operation process, radiation dose to the operator (group A) was 0.10 ± 0.03 (0.07-0.15) µSv, group B was 12.09 ± 4.67 (10-20) µSv. a difference that was found to be statistically significant (P < 0.001) between group A and group B. CONCLUSION: New cement delivery device plus meticulous preoperative preparation can significantly decrease radiation dose to operators.

An Extremely Simple Algorithm for Source Domain Reconstruction.

The aim of unsupervised domain adaptation (UDA) is to utilize knowledge from a source domain to enhance the performance of a given target domain. Due to the lack of accessibility to the target domain's labels, UDA's efficacy is highly reliant on the source domain's quality. However, it is often impractical and expensive to obtain an appropriate transferable source domain. To address this issue, we propose a novel UDA setting, source domain reconstruction (SDR), which seeks to construct a new transferable source domain utilizing labeled source samples and unlabeled target samples. SDR has a significant advantage over the conventional method as it is much less expensive to construct a suitable pseudo-source domain rather than collecting an actual transferable source domain in real-world scenarios. To test the practice of SDR, we investigate SDR theoretically. We propose an easily implementable algorithm, the domain MixUp (DMU), which is motivated by the MixUp strategy, to solve the SDR problem. The algorithm can be used to design a UDA framework to significantly enhance the performance of several existing UDA algorithms. Results from extensive experiments conducted on seven benchmarks (66 UDA tasks) indicate that the reconstructed source domain has stronger transferability than the original source domain.

Multidomain Adaptation With Sample and Source Distillation.

Unsupervised multidomain adaptation attracts increasing attention as it delivers richer information when tackling a target task from an unlabeled target domain by leveraging the knowledge attained from labeled source domains. However, it is the quality of training samples, not just the quantity, that influences transfer performance. In this article, we propose a multidomain adaptation method with sample and source distillation (SSD), which develops a two-step selective strategy to distill source samples and define the importance of source domains. To distill samples, the pseudo-labeled target domain is constructed to learn a series of category classifiers to identify transfer and inefficient source samples. To rank domains, the agreements of accepting a target sample as the insider of source domains are estimated by constructing a domain discriminator based on selected transfer source samples. Using the selected samples and ranked domains, transfer from source domains to the target domain is achieved by adapting multilevel distributions in a latent feature space. Furthermore, to explore more usable target information which is expected to enhance the performance across domains of source predictors, an enhancement mechanism is built by matching selected pseudo-labeled and unlabeled target samples. The degrees of acceptance learned by the domain discriminator are finally employed as source merging weights to predict the target task. Superiority of the proposed SSD is validated on real-world visual classification tasks.

TS-DM: A Time Segmentation-Based Data Stream Learning Method for Concept Drift Adaptation.

Concept drift arises from the uncertainty of data distribution over time and is common in data stream. While numerous methods have been developed to assist machine learning models in adapting to such changeable data, the problem of improperly keeping or discarding data samples remains. This may results in the loss of valuable knowledge that could be utilized in subsequent time points, ultimately affecting the model's accuracy. To address this issue, a novel method called time segmentation-based data stream learning method (TS-DM) is developed to help segment and learn the streaming data for concept drift adaptation. First, a chunk-based segmentation strategy is given to segment normal and drift chunks. Building upon this, a chunk-based evolving segmentation (CES) strategy is proposed to mine and segment the data chunk when both old and new concepts coexist. Furthermore, a warning level data segmentation process (CES-W) and a high-low-drift tradeoff handling process are developed to enhance the generalization and robustness. To evaluate the performance and efficiency of our proposed method, we conduct experiments on both synthetic and real-world datasets. By comparing the results with several state-of-the-art data stream learning methods, the experimental findings demonstrate the efficiency of the proposed method.

Deep Reinforcement Learning in Nonstationary Environments With Unknown Change Points.

Deep reinforcement learning (DRL) is a powerful tool for learning from interactions within a stationary environment where state transition and reward distributions remain constant throughout the process. Addressing the practical but challenging nonstationary environments with time-varying state transition or reward function changes during the interactions, ingenious solutions are essential for the stability and robustness of DRL agents. A key assumption to cope with nonstationary environments is that the change points between the previous and the new environments are known beforehand. Unfortunately, this assumption is impractical in many cases, such as outdoor robots and online recommendations. To address this problem, this article presents a robust DRL algorithm for nonstationary environments with unknown change points. The algorithm actively detects change points by monitoring the joint distribution of states and actions. A detection boosted, gradient-constrained optimization method then adapts the training of the current policy with the supporting knowledge of formerly well-trained policies. The previous policies and experience help the current policy adapt rapidly to environmental changes. Experiments show that the proposed method accumulates the highest reward among several alternatives and is the fastest to adapt to new environments. This work has compelling potential for increasing the environmental suitability of intelligent agents, such as drones, autonomous vehicles, and underwater robots.

Multiview Classification Through Learning From Interval-Valued Data.

The classification problem concerning crisp-valued data has been well resolved. However, interval-valued data, where all of the observations' features are described by intervals, are also a common data type in real-world scenarios. For example, the data extracted by many measuring devices are not exact numbers but intervals. In this article, we focus on a highly challenging problem called learning from interval-valued data (LIND), where we aim to learn a classifier with high performance on interval-valued observations. First, we obtain the estimation error bound of the LIND problem based on the Rademacher complexity. Then, we give the theoretical analysis to show the strengths of multiview learning on classification problems, which inspires us to construct a new algorithm called multiview interval information extraction (Mv-IIE) approach for improving classification accuracy on interval-valued data. The experiment comparisons with several baselines on both synthetic and real-world datasets illustrate the superiority of the proposed framework in handling interval-valued data. Moreover, we describe an application of Mv-IIE that we can prevent data privacy leakage by transforming crisp-valued (raw) data into interval-valued data.

Disentangling Stochastic PDE Dynamics for Unsupervised Video Prediction.

Unsupervised video prediction aims to predict future outcomes based on the observed video frames, thus removing the need for supervisory annotations. This research task has been argued as a key component of intelligent decision-making systems, as it presents the potential capacities of modeling the underlying patterns of videos. Essentially, the challenge of video prediction is to effectively model the complex spatiotemporal and often uncertain dynamics of high-dimensional video data. In this context, an appealing way of modeling spatiotemporal dynamics is to explore prior physical knowledge, such as partial differential equations (PDEs). In this article, considering real-world video data as a partly observed stochastic environment, we introduce a new stochastic PDE predictor (SPDE-predictor), which models the spatiotemporal dynamics by approximating a generalized form of PDEs while dealing with the stochasticity. A second contribution is that we disentangle the high-dimensional video prediction into low-level dimensional factors of variations: time-varying stochastic PDE dynamics and time-invariant content factors. Extensive experiments on four various video datasets show that SPDE video prediction model (SPDE-VP) outperforms both deterministic and stochastic state-of-the-art methods. Ablation studies highlight our superiority driven by both PDE dynamics modeling and disentangled representation learning and their relevance in long-term video prediction.

Learning Data Streams With Changing Distributions and Temporal Dependency.

In a data stream, concept drift refers to unpredictable distribution changes over time, which violates the identical-distribution assumption required by conventional machine learning methods. Current concept drift adaptation techniques mostly focus on a data stream with changing distributions. However, since each variable of a data stream is a time series, these variables normally have temporal dependency problems in the real world. How to solve concept drift and temporal dependency problems at the same time is rarely discussed in the concept-drift literature. To solve this situation, this article proves and validates that the testing error decreases faster if a predictor is trained on a temporally reconstructed space when drift occurs. Based on this theory, a novel drift adaptation regression (DAR) framework is designed to predict the label variable for data streams with concept drift and temporal dependency. A new statistic called local drift degree (LDD+) is proposed and used as a drift adaptation technique in the DAR framework to discard outdated instances in a timely way, thereby guaranteeing that the most relevant instances will be selected during the training process. The performance of DAR is demonstrated by a set of experimental evaluations on both synthetic data and real-world data streams.

Semi-Supervised Heterogeneous Domain Adaptation: Theory and Algorithms.

Semi-supervised heterogeneous domain adaptation (SsHeDA) aims to train a classifier for the target domain, in which only unlabeled and a small number of labeled data are available. This is done by leveraging knowledge acquired from a heterogeneous source domain. From algorithmic perspectives, several methods have been proposed to solve the SsHeDA problem; yet there is still no theoretical foundation to explain the nature of the SsHeDA problem or to guide new and better solutions. Motivated by compatibility condition in semi-supervised probably approximately correct (PAC) theory, we explain the SsHeDA problem by proving its generalization error - that is, why labeled heterogeneous source data and unlabeled target data help to reduce the target risk. Guided by our theory, we devise two algorithms as proof of concept. One, kernel heterogeneous domain alignment (KHDA), is a kernel-based algorithm; the other, joint mean embedding alignment (JMEA), is a neural network-based algorithm. When a dataset is small, KHDA's training time is less than JMEA's. When a dataset is large, JMEA is more accurate in the target domain. Comprehensive experiments with image/text classification tasks show KHDA to be the most accurate among all non-neural network baselines, and JMEA to be the most accurate among all baselines.

Evolving Gradient Boost: A Pruning Scheme Based on Loss Improvement Ratio for Learning Under Concept Drift.

In nonstationary environments, data distributions can change over time. This phenomenon is known as concept drift, and the related models need to adapt if they are to remain accurate. With gradient boosting (GB) ensemble models, selecting which weak learners to keep/prune to maintain model accuracy under concept drift is nontrivial research. Unlike existing models such as AdaBoost, which can directly compare weak learners' performance by their accuracy (a metric between [0, 1]), in GB, weak learners' performance is measured with different scales. To address the performance measurement scaling issue, we propose a novel criterion to evaluate weak learners in GB models, called the loss improvement ratio (LIR). Based on LIR, we develop two pruning strategies: 1) naive pruning (NP), which simply deletes all learners with increasing loss and 2) statistical pruning (SP), which removes learners if their loss increase meets a significance threshold. We also devise a scheme to dynamically switch between NP and SP to achieve the best performance. We implement the scheme as a concept drift learning algorithm, called evolving gradient boost (LIR-eGB). On average, LIR-eGB delivered the best performance against state-of-the-art methods on both stationary and nonstationary data.

Heterogeneous Multidomain Recommender System Through Adversarial Learning.

To solve the user data sparsity problem, which is the main issue in generating user preference prediction, cross-domain recommender systems transfer knowledge from one source domain with dense data to assist recommendation tasks in the target domain with sparse data. However, data are usually sparsely scattered in multiple possible source domains, and in each domain (source/target) the data may be heterogeneous, thus it is difficult for existing cross-domain recommender systems to find one source domain with dense data from multiple domains. In this way, they fail to deal with data sparsity problems in the target domain and cannot provide an accurate recommendation. In this article, we propose a novel multidomain recommender system (called HMRec) to deal with two challenging issues: 1) how to exploit valuable information from multiple source domains when no single source domain is sufficient and 2) how to ensure positive transfer from heterogeneous data in source domains with different feature spaces. In HMRec, domain-shared and domain-specific features are extracted to enable the knowledge transfer between multiple heterogeneous source and target domains. To ensure positive transfer, the domain-shared subspaces from multiple domains are maximally matched by a multiclass domain discriminator in an adversarial learning process. The recommendation in the target domain is completed by a matrix factorization module with aligned latent features from both the user and the item side. Extensive experiments on four cross-domain recommendation tasks with real-world datasets demonstrate that HMRec can effectively transfer knowledge from multiple heterogeneous domains collaboratively to increase the rating prediction accuracy in the target domain and significantly outperforms six state-of-the-art non-transfer or cross-domain baselines.

Bridging the Theoretical Bound and Deep Algorithms for Open Set Domain Adaptation.

In the unsupervised open set domain adaptation (UOSDA), the target domain contains unknown classes that are not observed in the source domain. Researchers in this area aim to train a classifier to accurately: 1) recognize unknown target data (data with unknown classes) and 2) classify other target data. To achieve this aim, a previous study has proven an upper bound of the target-domain risk, and the open set difference, as an important term in the upper bound, is used to measure the risk on unknown target data. By minimizing the upper bound, a shallow classifier can be trained to achieve the aim. However, if the classifier is very flexible [e.g., deep neural networks (DNNs)], the open set difference will converge to a negative value when minimizing the upper bound, which causes an issue where most target data are recognized as unknown data. To address this issue, we propose a new upper bound of target-domain risk for UOSDA, which includes four terms: source-domain risk, ϵ -open set difference ( ∆ϵ ), distributional discrepancy between domains, and a constant. Compared with the open set difference, ∆ϵ is more robust against the issue when it is being minimized, and thus we are able to use very flexible classifiers (i.e., DNNs). Then, we propose a new principle-guided deep UOSDA method that trains DNNs via minimizing the new upper bound. Specifically, source-domain risk and ∆ϵ are minimized by gradient descent, and the distributional discrepancy is minimized via a novel open set conditional adversarial training strategy. Finally, compared with the existing shallow and deep UOSDA methods, our method shows the state-of-the-art performance on several benchmark datasets, including digit recognition [modified National Institute of Standards and Technology database (MNIST), the Street View House Number (SVHN), U.S. Postal Service (USPS)], object recognition (Office-31, Office-Home), and face recognition [pose, illumination, and expression (PIE)].

Fast treatment and recycling method of large-scale vegetable wastes.

With rapid development of vegetable industry in China, in process of refrigerated transportation and storage, large-scale abandoned vegetable wastes (VW) need to be urgently treated alone since they rot very fast and would pollute the environment seriously. Existing treatment projects generally regard VW as garbage with high content of water and adopt the process of squeeze and sewage treatment, which leads to not only high treatment costs but also great resource waste. Therefore, according to the composition and degradation characteristics of VW, a novel fast treatment and recycling method of VW was proposed in this paper. VW are first degraded with thermostatic anaerobic digestion (AD) and then the residues decompose rapidly with thermostatic aerobic digestion to meet the farmland application standard. To verify the feasibility of the method, the pressed VW water (PVW) and VW from the VW treatment plant were mixed and degraded in two 0.56 m3 digesters, and degraded substances were continuously measured in 30 days' mesophilic AD at 37 ± 1 °C. Subsequently, the biogas slurry (BS) produced by AD is decomposed by thermostatic aerobic aeration decomposition at 30 °C for 48 h to rapidly decompose. BS was confirmed to use safely for plants by germination index (GI) test. The results show that 96 % chemical oxygen demand (COD) from 15,711 mg/L to 1000 mg/L within 31 days and the GI of treated BS was 81.75 %. Besides, nutrient elements of N, P, and K keep good abundance, no heavy metals, pesticide residue, and hazardous substances were found. Other parameters were all lower than the BS placed for a half-year. VW are fast-treated and recycled with the new method, which provides a novel method for fast treatment and recycling of large-scale VW.