Search

Mycotoxin contamination is a globally concerned problem for food and agricultural products since it may directly or indirectly induce severe threats to human health. Sensitive and selective screening is an efficient strategy to prevent or reduce human and animal exposure to mycotoxins. However, enormous challenges exist in the determination of mycotoxins, arising from complex sample matrices, trace-level analytes, and the co-occurrence of diverse mycotoxins. Appropriate sample preparation is essential to isolate, purify, and enrich mycotoxins from complicated matrices, thus decreasing sample matrix effects and lowering detection limits. With the cross-disciplinary development, new solid-phase extraction strategies have been exploited and integrated with nanotechnology to meet the challenges of mycotoxin analysis. This review summarizes the advance and progress of solid-phase extraction techniques as the methodological solutions for mycotoxin analysis. Emphases are paid on nanomaterials fabricated as trapping media of solid-phase extraction techniques, including carbonaceous nanoparticles, metal/metal oxide-based nanoparticles, and nanoporous materials. Advantages and limitations are discussed, along with the potential prospects.

Phthalate esters (PAEs) are ubiquitous pollutants in the environment with toxicological and epidemiological effects for humans. As one of the daily necessities, edible plant oil is an important exposure source of PAEs, due to the inevitable contact with PAE-containing materials and the intrinsic lipid solubility of PAEs. However, limited information is currently available on the exposure risk of PAEs in commercial plant oil. This study was aimed at investigating the occurrence and risk assessment of PAEs in plant oils with a high-frequency import rate in west China. The analysis method was referenced to the Chinese national standard for the determination of PAEs in food. Results indicated that PAEs (mainly including DBP and DEHP) were ubiquitous contaminants in imported plant oils with the detectable rate being up to 56.83% in 366 samples. The detected concentrations were in the range of 0.10-3.20 mg kg

Magnetic solid-phase extraction (MSPE) strategy based on the Fe

The meta-heuristic is becoming important in the field of modern optimization, and gained more and more attention. In the past thirty years there has been a wide range of meta-heuristic, such as gravitational search algorithm (GSA), has achieved a great success. Spherical search (SS) is the one of newest proposed meta-heuristic algorithms. SS performs search effectively in exploration, but due to the lack of local exploitation ability, it converges slowly and can’t exploit the small region around the current promising solution. This paper proposes a novel optimization algorithm, namely SSGSA, which is inherited from the SS and GSA to combine the effective exploration and exploitation of each algorithm, respectively. To evaluate the effectiveness of SSGSA, we compared it with the original SS, original GSA, particle swarm optimization, and whale optimization algorithm on the IEEE CEC’17 benchmark function suit. Experimental results show that the proposed new method outperforms its competitors in terms of convergence speed and solution accuracy.

The Fighting Game AI Competition (FTGAIC) provides a challenging benchmark for 2-player video game AI. The challenge arises from the large action space, diverse styles of characters and abilities, and the real-time nature of the game. In this paper, we propose a novel algorithm that combines Rolling Horizon Evolution Algorithm (RHEA) with opponent model learning. The approach is readily applicable to any 2-player video game. In contrast to conventional RHEA, an opponent model is proposed and is optimized by supervised learning with cross-entropy and reinforcement learning with policy gradient and Q-learning respectively, based on history observations from opponent. The model is learned during the live gameplay. With the learned opponent model, the extended RHEA is able to make more realistic plans based on what the opponent is likely to do. This tends to lead to better results. We compared our approach directly with the bots from the FTGAIC 2018 competition, and found our method to significantly outperform all of them, for all three character. Furthermore, our proposed bot with the policy-gradient-based opponent model is the only one without using Monte-Carlo Tree Search (MCTS) among top five bots in the 2019 competition in which it achieved second place, while using much less domain knowledge than the winner., Comment: 10 pages, 7 figures

After artificial intelligent (AI) scientists have conquered Go game, StarCraft has been the next biggest challenge. A highly intelligent AI system that is able to beat human professional players is expected. In this paper, we review the recent development of computational intelligence (CI) in the field of StarCraft AI. Successful applications of CI techniques are analyzed and compared from different levels of AI functionality. It should be noted that current StarCraft AI is highly dependent on human experience and is far away from a completely intelligent one. New frameworks and techniques are still expected to improve the intelligence.

At present, there is a growing concern about human quality of life. In particular, there is an awareness of the impact of volatile organic compounds (VOCs) on the environment and human health, so the monitoring of human exposure to VOCs is an increasingly urgent need. Biomonitoring is theoretically more accurate compared with traditional ambient air monitoring, and it plays an essential role in human environmental exposure assessment. Breath analysis is a biomonitoring method with many advantages, which is applicable to assessments of human exposure to a large number of VOCs. Techniques are being developed to improve the sensitivity and precision of breath analysis based on in-direct and direct measurements which will be reviewed in this paper. This paper briefly reviews the frequently used methods in both of these categories, specifically highlighting some promising new techniques. Furthermore, this review also provides theoretical background knowledge about the use of breath analysis as a biomonitoring tool for human exposure assessment. A review of the application of breath analysis to human exposure monitoring during last two decades is also provided according to occupational/non-occupational exposure. Obstacles and potential challenges in this field are also summarized. Based on the gradual improvements in the theoretical basis and technology reviewed in this paper, breath analysis is an enormous potential approach for the monitoring of human exposure to VOCs.

StarCraft II is one of the most popular real-time strategy games and has become an important benchmark for AI research as it provides a complex environment with numerous challenges. The build order problem is one of the key challenges which concern the order and type of buildings and units to produce based on current game situation. In contrast to existing hand-craft methods, we propose two reinforcement learning based models: Neural Network Fitted Q-Learning (NNFQ) and Convolutional Neural Network Fitted Q-Learning (CNNFQ). NNFQ and CNNFQ have been applied into a simple bot for fighting against the enemy race. Experimental results show that both these two models are capable of finding the most effective production sequence to defeat the opponent.

A kind of porous ionic liquid polymer was designed as coating material for solid-phase microextraction (SPME) fibers to extract polar organic acids. The synthesized ionic liquids (1-vinyl-3-(4-vinyl-benzyl)imidazolium chloride) were polymerized in the presence of azobisisobutyronitrile (AIBN), and subsequently coated on stainless steel wires with physical adhesion. To guarantee the adhesion between coating and the surface of stainless steel wires, micro-wave induced plasma was applied to modify the stainless steel wires. The derivatives of extracted organic acids were analyzed by gas chromatography combined with mass spectrometry (GC-MS) with N-tert-Butyldimethylsilyl-N-methyltrifluoroacetamide (MTBSTFA) as the derivatization reagent. Under the optimal condition, a calibration study was carried out to evaluate the performance of the SPME fibers. The calibration curves were linear (R2>0.99) in the range from 0.01 μg/ml to 1 μg/ml for six organic acids. Detection limits were detected down to 0.07 ng/ml. To further validate the extraction efficiency of the fibers, experiments were conducted in comparison with commercial fibers. Lower LOD and LOQ values were obtained by the in-house fabricated fibers especially for polyatomic acids. Good repeatability (RSD≤16%) and fiber-to-fiber reproducibility (RSD≤20%) were obtained. The acceptable recovery of the spiked grape wine samples ranged from 78.19% to 98.11%. Additionally, long lifetime and good durability of the fibers have been demonstrated. The performance of the established SPME-GC-MS method for analysis of organic acids was shown to be greatly improved in comparison to the direct HPLC/CE method or other SPME materials as reported in the literature. In summary, a feasible and effective approach has been demonstrated to fabricate the in-house fabricated fibers based on porous ionic liquid polymers, which were proved to be advantageous and sensitive for extraction of polar organic acids. Simultaneously, application of the SPME derivatization GC-MS method for analysis of organic acids has been successfully verified as feasible and reliable.

Most agents obtain knowledges from natural scenes through some single preestablished rules. In practice, those single rules can’t achieve the aim to freely awareness the natural scenes, such as the visual scenes. Inspired by biological visual cortex (V1) and higher brain areas perceiving visual features, in this paper we propose an improved visual awareness module, called as visual scenes mining module, for the agent ABGP-CNN in order to directly mine the visual scenes information. Then ABGP-CNN with the visual scenes mining module is deployed on a toy car. The visual information mining from the nature scenes is served as the knowledges of the agent ABGP-CNN to drive the toy car. The toy car deployed the agent ABGP-CNN can easily understand the special natural visual scenes, and has the ability to plan its behaviors according to the visual information mining from the nature scenes. The application of the agent ABGP-CNN with visual scenes mining module enhances the capability of communication between the toy car and the natural visual scenes.

The Gomoku board game is a longstanding challenge for artificial intelligence research. With the development of deep learning, move prediction can help to promote the intelligence of board game agents as proven in AlphaGo. Following this idea, we train deep convolutional neural networks by supervised learning to predict the moves made by expert Gomoku players from RenjuNet dataset. We put forward a number of deep neural networks with different architectures and different hyperparameters to solve this problem. With only the board state as the input, the proposed deep convolutional neural networks are able to recognize some special features of Gomoku and select the most likely next move. The final neural network achieves the accuracy of move prediction of about 42% on the RenjuNet dataset, which reaches the level of expert Gomoku players. In addition, it is promising to generate strong Gomoku agents of human-level with the move prediction as a guide.

Integral analysis plays an important role in study and quality control of substances with complex matrices in our daily life. As the preliminary construction of integral analysis of substances with complex matrices, developing a relatively comprehensive and sensitive methodology might offer more informative and reliable characteristic components. Flavoring mixtures belonging to the representatives of substances with complex matrices have now been widely used in various fields. To better study and control the quality of flavoring mixtures as additives in food industry, an in-house fabricated solid-phase microextraction (SPME) fiber was prepared based on sol-gel technology in this work. The active organic component of the fiber coating was multi-walled carbon nanotubes (MWCNTs) functionalized with hydroxyl-terminated polydimethyldiphenylsiloxane, which integrate the non-polar and polar chains of both materials. In this way, more sensitive extraction capability for a wider range of compounds can be obtained in comparison with commercial SPME fibers. Preliminarily integral analysis of three similar types of samples were realized by the optimized SPME-GC-MS method. With the obtained GC-MS data, a valid and well-fit model was established by partial least square discriminant analysis (PLS-DA) for classification of these samples (R2X=0.661, R2Y=0.996, Q2=0.986). The validity of the model (R2=0.266, Q2=-0.465) has also approved the potential to predict the "belongingness" of new samples. With the PLS-DA and SPSS method, further screening out the markers among three similar batches of samples may be helpful for monitoring and controlling the quality of the flavoring mixtures as additives in food industry. Conversely, the reliability and effectiveness of the GC-MS data has verified the comprehensive and efficient extraction performance of the in-house fabricated fiber.

Breath analysis is a noninvasive diagnostic method that profiles a person's physical state by volatile organic compounds in the breath. It has huge potential in the field of disease diagnosis. In order to offer opportunities for practical applications, various GC-based techniques have been investigated for on-line breath analysis since GC is the most preferred technique for mixed gas separation. This article reviews the development of breath analysis and GC-based techniques in basic breath research, involving sampling methods, preconcentration methods, conventional GC-based techniques, and newly developed GC techniques for breath analysis. The combination of GC and newly developed detection techniques takes advantages of the virtues of each. In addition, portable GC or micro GC are poised to become field GC-based techniques in breath analysis. Challenges faced in GC-based techniques for breath analysis are discussed candidly. Effective cooperation of experts from different fields is urgent to promote the development of breath analysis.

In this work, a laboratory preparation method based on sol-gel technology was proposed to develop a new kind of SPME (solid phase microextraction) fibers. Multi-walled carbon nanotubes (MWCNT) were selected as sol-gel active organic component. Stainless steel wires were used as the substrate of the fibers. Instead of traditional modification methods, microwave induced plasma was used to modify the stainless steel wire surface, resulting in a significant improvement in chemical adhesion of the fiber substrate and coating. The MWCNT coating exhibited several good properties. Acceptable fiber-to-fiber reproducibility (RSD≤13%) and repeatability (RSD

Breast cancer (BC) remains the most commonly diagnosed malignancy in women. We investigated 4 straight aldehydes in the exhaled breath as potential early BC diagnostic biomarkers.End-tailed breath were collected by Bio-VOC® sampler and assayed by gas chromatography-mass spectrometry. Kruskal-Wallis one-way analysis of variance test and binary logistic regression were used for data analysis. The diagnostic accuracies were evaluated by receiver operating characteristic curves. A predictive model/equation was generated using the 4 biomarkers and validated by leave-one-out cross-validation.All four potential biomarkers demonstrated significant differences in concentrations between BC and healthy controls (HC) (p0.05). The areas under the curves (AUCs) in HC vs BCI-II model using hexanal, heptanal, octanal, and nonanal were 0.816, 0.809, 0.731, and 0.830, respectively. The AUC for their combined use was 0.934 (sensitivity 91.7%, specificity 95.8%) in the early diagnosis of BC. The predictive model/equation exhibited good sensitivity (72.7%) and specificity (91.7%) in distinguishing between HC and BC (cross-validation: sensitivity 68.2% and specificity 91.7%).The diagnostic values of 4 exhaled straight aldehydes as early diagnostic biomarkers for BC were successfully verified and the diagnostic accuracy improved in their combined use.