Crop cultivation planning with fuzzy estimation using water wave optimization.

In a complex agricultural region, determine the appropriate crop for each plot of land to maximize the expected total profit is the key problem in cultivation management. However, many factors such as cost, yield, and selling price are typically uncertain, which causes an exact programming method impractical. In this paper, we present a problem of crop cultivation planning, where the uncertain factors are estimated as fuzzy parameters. We adapt an efficient evolutionary algorithm, water wave optimization (WWO), to solve this problem, where each solution is evaluated based on three metrics including the expected, optimistic and pessimistic values, the combination of which enables the algorithm to search credible solutions under uncertain conditions. Test results on a set of agricultural regions in East China showed that the solutions of our fuzzy optimization approach obtained significantly higher profits than those of non-fuzzy optimization methods based on only the expected values.

Predicting Demands of COVID-19 Prevention and Control Materials via Co-Evolutionary Transfer Learning.

The novel coronavirus pneumonia (COVID-19) has created great demands for medical resources. Determining these demands timely and accurately is critically important for the prevention and control of the pandemic. However, even if the infection rate has been estimated, the demands of many medical materials are still difficult to estimate due to their complex relationships with the infection rate and insufficient historical data. To alleviate the difficulties, we propose a co-evolutionary transfer learning (CETL) method for predicting the demands of a set of medical materials, which is important in COVID-19 prevention and control. CETL reuses material demand knowledge not only from other epidemics, such as severe acute respiratory syndrome (SARS) and bird flu but also from natural and manmade disasters. The knowledge or data of these related tasks can also be relatively few and imbalanced. In CETL, each prediction task is implemented by a fuzzy deep contractive autoencoder (CAE), and all prediction networks are cooperatively evolved, simultaneously using intrapopulation evolution to learn task-specific knowledge in each domain and using interpopulation evolution to learn common knowledge shared across the domains. Experimental results show that CETL achieves high prediction accuracies compared to selected state-of-the-art transfer learning and multitask learning models on datasets during two stages of COVID-19 spreading in China.

Integration of machine learning prediction and heuristic optimization for mask delivery in COVID-19.

The novel coronavirus pneumonia (COVID-19) has created huge demands for medical masks that need to be delivered to a lot of demand points to protect citizens. The efficiency of delivery is critical to the prevention and control of the epidemic. However, the huge demands for masks and massive number of demand points scattered make the problem highly complex. Moreover, the actual demands are often obtained late, and hence the time duration for solution calculation and mask delivery is often very limited. Based on our practical experience of medical mask delivery in response to COVID-19 in China, we present a hybrid machine learning and heuristic optimization method, which uses a deep learning model to predict the demand of each region, schedules first-echelon vehicles to pre-distribute the predicted number of masks from depot(s) to regional facilities in advance, reassigns demand points among different regions to balance the deviations of predicted demands from actual demands, and finally routes second-echelon vehicles to efficiently deliver masks to the demand points in each region. For the subproblems of demand point reassignment and two-batch routing whose complexities are significantly lower, we propose variable neighborhood tabu search heuristics to efficiently solve them. Application of the proposed method in emergency mask delivery in three megacities in China during the peak of COVID-19 demonstrated its significant performance advantages over other methods without pre-distribution or reassignment. We also discuss key success factors and lessons learned to facilitate the extension of our method to a wider range of problems.

A robust multi-objective model for healthcare resource management and location planning during pandemics.

In this study, we consider the problem of healthcare resource management and location planning problem during the early stages of a pandemic/epidemic under demand uncertainty. Our main ambition is to improve the preparedness level and response effectiveness of healthcare authorities in fighting pandemics/epidemics by implementing analytical techniques. Building on lessons from the Chinese experience in the COVID-19 outbreak, we first develop a deterministic multi-objective mixed integer linear program (MILP) which determines the location and size of new pandemic hospitals (strategic level planning), periodic regional health resource re-allocations (tactical level planning) and daily patient-hospital assignments (operational level planning). Taking the forecasted number of cases along a planning horizon as an input, the model minimizes the weighted sum of the number of rejected patients, total travel distance, and installation cost of hospitals subject to real-world constraints and organizational rules. Next, accounting for the uncertainty in the spread speed of the disease, we employ an across scenario robust (ASR) model and reformulate the robust counterpart of the deterministic MILP. The ASR attains relatively more realistic solutions by considering multiple scenarios simultaneously while ensuring a predefined threshold of relative regret for the individual scenarios. Finally, we demonstrate the performance of proposed models on the case of Wuhan, China. Taking the 51 days worth of confirmed COVID-19 case data as an input, we solve both deterministic and robust models and discuss the impact of all three level decisions to the quality and performance of healthcare services during the pandemic. Our case study results show that although it is a challenging task to make strategic level decisions based on uncertain forecasted data, an immediate action can considerably improve the response effectiveness of healthcare authorities. Another important observation is that, the installation times of pandemic hospitals have significant impact on the system performance in fighting with the shortage of beds and facilities.

Multi-Objective Optimization of Integrated Civilian-Military Scheduling of Medical Supplies for Epidemic Prevention and Control.

In a large-scale epidemic, such as the novel coronavirus pneumonia (COVID-19), there is huge demand for a variety of medical supplies, such as medical masks, ventilators, and sickbeds. Resources from civilian medical services are often not sufficient for fully satisfying all of these demands. Resources from military medical services, which are normally reserved for military use, can be an effective supplement to these demands. In this paper, we formulate a problem of integrated civilian-military scheduling of medical supplies for epidemic prevention and control, the aim of which is to simultaneously maximize the overall satisfaction rate of the medical supplies and minimize the total scheduling cost, while keeping a minimum ratio of medical supplies reservation for military use. We propose a multi-objective water wave optimization (WWO) algorithm in order to efficiently solve this problem. Computational results on a set of problem instances constructed based on real COVID-19 data demonstrate the effectiveness of the proposed method.

Tridirectional Transfer Learning for Predicting Gastric Cancer Morbidity.

Our previous study has constructed a deep learning model for predicting gastrointestinal infection morbidity based on environmental pollutant indicators in some regions in central China. This article aims to adapt the prediction model for three purposes: 1) predicting the morbidity of a different disease in the same region; 2) predicting the morbidity of the same disease in a different region; and 3) predicting the morbidity of a different disease in a different region. We propose a tridirectional transfer learning approach, which achieves the abovementioned three purposes by: 1) developing a combined univariate regression and multivariate Gaussian model for establishing the relationship between the morbidity of the target disease and that of the source disease together with the high-level pollutant features in the current source region; 2) using mapping-based deep transfer learning to extend the current model to predict the morbidity of the source disease in both source and target regions; and 3) applying the pattern of the combined model in the source region to the extended model to derive a new combined model for predicting the morbidity of the target disease in the target region. We select gastric cancer as the target disease and use the proposed transfer learning approach to predict its morbidity in the source region and three target regions. The results show that, given only a limited number of labeled samples, our approach achieves an average prediction accuracy of over 80% in the source region and up to 78% in the target regions, which can contribute considerably to improving medical preparedness and response.

Real-time neural network scheduling of emergency medical mask production during COVID-19.

During the outbreak of the novel coronavirus pneumonia (COVID-19), there is a huge demand for medical masks. A mask manufacturer often receives a large amount of orders that must be processed within a short response time. It is of critical importance for the manufacturer to schedule and reschedule mask production tasks as efficiently as possible. However, when the number of tasks is large, most existing scheduling algorithms require very long computational time and, therefore, cannot meet the needs of emergency response. In this paper, we propose an end-to-end neural network, which takes a sequence of production tasks as inputs and produces a schedule of tasks in a real-time manner. The network is trained by reinforcement learning using the negative total tardiness as the reward signal. We applied the proposed approach to schedule emergency production tasks for a medical mask manufacturer during the peak of COVID-19 in China. Computational results show that the neural network scheduler can solve problem instances with hundreds of tasks within seconds. The objective function value obtained by the neural network scheduler is significantly better than those of existing constructive heuristics, and is close to those of the state-of-the-art metaheuristics whose computational time is unaffordable in practice.

Quarantine Vehicle Scheduling for Transferring High-Risk Individuals in Epidemic Areas.

In a large-scale epidemic outbreak, there can be many high-risk individuals to be transferred for medical isolation in epidemic areas. Typically, the individuals are scattered across different locations, and available quarantine vehicles are limited. Therefore, it is challenging to efficiently schedule the vehicles to transfer the individuals to isolated regions to control the spread of the epidemic. In this paper, we formulate such a quarantine vehicle scheduling problem for high-risk individual transfer, which is more difficult than most well-known vehicle routing problems. To efficiently solve this problem, we propose a hybrid algorithm based on the water wave optimization (WWO) metaheuristic and neighborhood search. The metaheuristic uses a small population to rapidly explore the solution space, and the neighborhood search uses a gradual strategy to improve the solution accuracy. Computational results demonstrate that the proposed algorithm significantly outperforms several existing algorithms and obtains high-quality solutions on real-world problem instances for high-risk individual transfer in Hangzhou, China, during the peak period of the novel coronavirus pneumonia (COVID-19).

Correlations Study Between 18F-FDG PET/CT Metabolic Parameters Predicting Epidermal Growth Factor Receptor Mutation Status and Prognosis in Lung Adenocarcinoma.

Purpose: This study assessed the ability of metabolic parameters from 18Fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) and clinicopathological data to predict epidermal growth factor receptor (EGFR) expression/mutation status in patients with lung adenocarcinoma and to develop a prognostic model based on differences in EGFR expression status, to enable individualized targeted molecular therapy. Patients and Methods: Metabolic parameters and clinicopathological data from 200 patients diagnosed with lung adenocarcinoma between July 2009 and November 2016, who underwent 18F-FDG PET/CT and EGFR mutation testing, were retrospectively evaluated. Multivariate logistic regression was applied to significant variables to establish a prediction model for EGFR mutation status. Overall survival for both mutant and wild-type EGFR was analyzed to establish a multifactor Cox regression model. Results: Of the 200 patients, 115 (58%) exhibited EGFR mutations and 85 (42%) were wild-type. Among selected metabolic parameters, metabolic tumor volume (MTV) demonstrated a significant difference between wild-type and mutant EGFR mutation status, with an area under the receiver operating characteristic curve (AUC) of 0.60, which increased to 0.70 after clinical data (smoking status) were combined. Survival analysis of wild-type and mutant EGFR yielded mean survival times of 34.451 (95% CI 28.654-40.249) and 53.714 (95% CI 44.331-63.098) months, respectively. Multivariate Cox regression revealed that mutation type, tumor stage, and thyroid transcription factor-1 (TTF-1) expression status were the main factors influencing patient prognosis. The hazard ratio for mutant EGFR was 0.511 (95% CI 0.303-0.862) times that of wild-type, and the risk of death was lower for mutant EGFR than for wild-type. The risk of death was lower in TTF-1-positive than in TTF-1-negative patients. Conclusion: 18F-FDG PET/CT metabolic parameters combined with clinicopathological data demonstrated moderate diagnostic efficacy in predicting EGFR mutation status and were associated with prognosis in mutant and wild-type EGFR non-small-cell lung cancer (NSCLC), thus providing a reference for individualized targeted molecular therapy.

Effects of Food Contamination on Gastrointestinal Morbidity: Comparison of Different Machine-Learning Methods.

Morbidity prediction can be useful in improving the effectiveness and efficiency of medical services, but accurate morbidity prediction is often difficult because of the complex relationships between diseases and their influencing factors. This study investigates the effects of food contamination on gastrointestinal-disease morbidities using eight different machine-learning models, including multiple linear regression, a shallow neural network, and three deep neural networks and their improved versions trained by an evolutionary algorithm. Experiments on the datasets from ten cities/counties in central China demonstrate that deep neural networks achieve significantly higher accuracy than classical linear-regression and shallow neural-network models, and the deep denoising autoencoder model with evolutionary learning exhibits the best prediction performance. The results also indicate that the prediction accuracies on acute gastrointestinal diseases are generally higher than those on other diseases, but the models are difficult to predict the morbidities of gastrointestinal tumors. This study demonstrates that evolutionary deep-learning models can be utilized to accurately predict the morbidities of most gastrointestinal diseases from food contamination, and this approach can be extended for the morbidity prediction of many other diseases.

An unprecedented pentanary chalcohalide with Mn atoms in two chemical environments: unique bonding characteristics and magnetic properties.

Mixed-anion chalcohalides have attracted significant attention lately, attributable to their unique structure compositions and captivating physicochemical properties. Herein, an unprecedented pentanary chalcohalide, Cs2[Mn2Ga3S7Cl] (1), was discovered by solid-state reaction at 1223 K. It is constructed by alternately stacked layers, each of which is made by a 2D [Ga3S9]9- ribbon embedded with 1D [Mn2S8Cl]13- chains. The coexistence of two Mn-coordinated polyhedra ([MnS6] octahedra and hetero-ligand [MnS3Cl] tetrahedra) in one material is surprisingly observed for the first time in the Mn-containing inorganic chalcogenides or chalcohalides. More interestingly, it exhibits ferrimagnetic (FIM) behaviour, which could be correlated to the magnetic sub-lattice sites with different coordination geometries. This work suggests a new route for designing and searching for unique functional chalcohalides with different chemical environments.

Generative adversarial network based telecom fraud detection at the receiving bank.

Recently telecom fraud has become a serious problem especially in developing countries such as China. At present, it can be very difficult to coordinate different agencies to prevent fraud completely. In this paper we study how to detect large transfers that are sent from victims deceived by fraudsters at the receiving bank. We propose a new generative adversarial network (GAN) based model to calculate for each large transfer a probability that it is fraudulent, such that the bank can take appropriate measures to prevent potential fraudsters to take the money if the probability exceeds a threshold. The inference model uses a deep denoising autoencoder to effectively learn the complex probabilistic relationship among the input features, and employs adversarial training that establishes a minimax game between a discriminator and a generator to accurately discriminate between positive samples and negative samples in the data distribution. We show that the model outperforms a set of well-known classification methods in experiments, and its applications in two commercial banks have reduced losses of about 10 million RMB in twelve weeks and significantly improved their business reputation.

Coexistence of Strong Second Harmonic Generation Response and Wide Band Gap in AZn4 Ga5 S12 (A=K, Rb, Cs) with 3D Diamond-like Frameworks.

Mid-infrared (MIR, 2-20 µm) second-order nonlinear optical (NLO) materials with outstanding performances are of great importance in laser science and technology. However, the enormous challenge to design and synthesize an excellent MIR NLO material lies in achieving simultaneously a strong second harmonic generation (SHG) response [dij >0.6 × AgGaS2 (AGS)] and wide band gap (Eg >3.5 eV). Herein three new MIR NLO materials, AZn4 Ga5 S12 (A=K, Rb, Cs) are reported, which crystallize in the KCd4 Ga5 S12 -type structure and adopt a 3D diamond-like framework (DLF) consisting of MS4 (M=Zn/Ga) tetrahedra; achieving the desired balance with strong powder SHG response (1.2-1.4 × AGS) and wide band gap (Eg ≈3.65 eV). Moreover, they also show large laser induced damage thresholds (LIDTs, 36 × AGS), a wide range of optical transparency (0.4-25 µm) and ultrahigh thermal stability (up to 1400 K). Upon analyzing the structure-property relationship of AXII4 XIII5 Q12 family, these 3D DLF structures can be used as a highly versatile and tunable platform for designing excellent MIR NLO materials.

Two excellent phase-matchable infrared nonlinear optical materials based on 3D diamond-like frameworks: RbGaSn2Se6 and RbInSn2Se6.

Mid- and far-infrared (MFIR) nonlinear optical (NLO) crystals with excellent performances are critical to laser frequency-conversion technology. However, the current commercial MFIR NLO crystals, including AgGaS2 (AGS), AgGaSe2 and ZnGeP2, suffer from certain intrinsic drawbacks and cannot achieve a good balance between large second-harmonic generation (SHG) efficiency and high laser-induced damage thresholds (LIDTs). Herein, we report two new phase-matchable MFIR NLO chalcogenides, specifically RbXSn2Se6 (X = Ga, In), which were successfully synthesized by high-temperature solid-state reactions. The remarkable structural feature of these materials was their 3D diamond-like framework (DLF) stacked by M3Se9 (M = X/Sn) asymmetric building units of vertex-sharing MSe4 tetrahedra along the c axis. Significantly, both of the materials showed the excellent NLO performances with the desired balance between their large SHG efficiencies (4.2 and 4.8 × benchmark AGS) and large LIDTs (8.9 and 8.1 × benchmark AGS), demonstrating that the title compounds meet the crucial conditions as promising MFIR NLO candidates. Furthermore, the crystal structures, synthesis, and theoretical analysis, as well as optical properties are presented herein.

Ba5Cu8In2S12: a quaternary semiconductor with a unique 3D copper-rich framework and ultralow thermal conductivity.

A novel quaternary sulfide, Ba5Cu8In2S12 (1), has been successfully synthesized via a high-temperature solid-state reaction. It contains Cu8S10S4/2 clusters as basic building blocks, which are connected to one another by discrete In3+ ions to generate a 3D copper-rich framework, where the Ba2+ cations reside. Interestingly, such large clusters that are fused by five crystallographically independent Cu sites with three different chemical environments result in the increase of phonon scattering, which is the crucial factor to the exceptionally low lattice thermal conductivity (ca. 0.28 W m-1 K-1 at 773 K) in 1.

Fireworks Algorithm with Enhanced Fireworks Interaction.

As a relatively new metaheuristic in swarm intelligence, fireworks algorithm (FWA) has exhibited promising performance on a wide range of optimization problems. This paper aims to improve FWA by enhancing fireworks interaction in three aspects: 1) Developing a new Gaussian mutation operator to make sparks learn from more exemplars; 2) Integrating the regular explosion operator of FWA with the migration operator of biogeography-based optimization (BBO) to increase information sharing; 3) Adopting a new population selection strategy that enables high-quality solutions to have high probabilities of entering the next generation without incurring high computational cost. The combination of the three strategies can significantly enhance fireworks interaction and thus improve solution diversity and suppress premature convergence. Numerical experiments on the CEC 2015 single-objective optimization test problems show the effectiveness of the proposed algorithm. The application to a high-speed train scheduling problem also demonstrates its feasibility in real-world optimization problems.

Tailored synthesis of nonlinear optical quaternary chalcohalides: Ba4Ge3S9Cl2, Ba4Si3Se9Cl2 and Ba4Ge3Se9Cl2.

Three novel zero-dimensional quaternary chalcohalides, Ba4Ge3S9Cl2, Ba4Si3Se9Cl2 and Ba4Ge3Se9Cl2, which crystallize in the polar noncentrosymmetric space group P63 (no. 173), have been rationally synthesized by a tailored approach on the basis of unique [M3Q9]6- (M = Ge, Q = S; M = Ge/Si, Q = Se) units with Ba2+ cations and Cl- anions occupying the interspaces. The [M3Q9]6- units which consist of three Q-corner-sharing [MQ4]4- tetrahedra, are arranged along the 63 screw axis. Remarkably, Ba4Ge3S9Cl2 exhibits a strong powder second harmonic generation (SHG) response that is 2.4 times that of benchmark AgGaS2 at a laser radiation of 2.05 µm in the same particle size range of 46-74 µm. Furthermore, theoretical studies based on the density functional theory helped to gain insight into the origin of the SHG.

Airline Passenger Profiling Based on Fuzzy Deep Machine Learning.

Passenger profiling plays a vital part of commercial aviation security, but classical methods become very inefficient in handling the rapidly increasing amounts of electronic records. This paper proposes a deep learning approach to passenger profiling. The center of our approach is a Pythagorean fuzzy deep Boltzmann machine (PFDBM), whose parameters are expressed by Pythagorean fuzzy numbers such that each neuron can learn how a feature affects the production of the correct output from both the positive and negative sides. We propose a hybrid algorithm combining a gradient-based method and an evolutionary algorithm for training the PFDBM. Based on the novel learning model, we develop a deep neural network (DNN) for classifying normal passengers and potential attackers, and further develop an integrated DNN for identifying group attackers whose individual features are insufficient to reveal the abnormality. Experiments on data sets from Air China show that our approach provides much higher learning ability and classification accuracy than existing profilers. It is expected that the fuzzy deep learning approach can be adapted for a variety of complex pattern analysis tasks.

Non-centrosymmetric Selenides AZn4 In5 Se12 (A=Rb, Cs): Synthesis, Characterization and Nonlinear Optical Properties.

Two new non-centrosymmetric polar quaternary selenides, namely, RbZn4 In5 Se12 and CsZn4 In5 Se12 , have been synthesized and structurally characterized. They exhibit a 3D diamond-like framework (DLF) consisting of corner-shared MSe4 (M=Zn/In) tetrahedra, in which the A+ ions are located. Both compounds are thermally stable up to 1300 K and exhibit large transmittance in the infrared region (0.65-25 µm) with measured optical band gaps of 2.06 eV for RbZn4 In5 Se12 and 2.11 eV for CsZn4 In5 Se12 . Inspiringly, they exhibit a good balance between strong second harmonic generation (SHG) efficiency (3.9 and 3.5×AgGaS2 ) and high laser-induced damage thresholds (13.0×AgGaS2 ). Theoretical calculations based on density functional theory (DFT) methods confirm that such strong SHG responses originate from the 3D DLF structure.

AXXTe12 (A = Rb, Cs; XII = Mn, Zn, Cd; XIII = Ga, In): quaternary semiconducting tellurides with very low thermal conductivities.

The discovery of novel materials with very low thermal conductivity is paramount to improving the efficiency of thermoelectric devices. Here we present a series of quaternary semiconducting tellurides AXXTe12 (A = Rb, Cs; XII = Mn, Zn, Cd; XIII = Ga, In) with three-dimensional (3D) diamond-like frameworks (DLFs) and they exhibit a very low thermal conductivity (ca. 0.26-0.42 W m-1 K-1) around 800 K.