Design and implementation of one-stop analysis and prediction platform for agricultural big data

With the continuous outbreak of COVID-19, it is becoming more and more difficult for individual farmers to survive only by offline production and sales. However, in this context, a perfect and humanized platform system has not yet been designed and put on the market. This system is committed to building a one-stop analysis and prediction platform for agricultural big data. Platform of the main service object for farmers, the use of the combination of online and offline model to analyze agricultural data, online business mainly for agriculture market dynamics, cost prediction, the profit prediction and disaster prediction, analysis of information related to supply reference for farmers and to cross-strait agricultural cooperation pilot zone of offline business provide the corresponding data of agricultural industry, To promote agricultural informatization and agricultural production, management, management, service for the service of effective integration between fulcrum, aiming at the current situation of agricultural development, agricultural products in rural areas are poor and on the development of the agricultural innovation, using the data from becoming more complete information and the foundation of agriculture information construction, to 'Internet + modern agriculture' as an opportunity to help farmers set up network information service. © 2022 IEEE.

Prediction-driven collaborative emergency medical resource allocation with deep learning and optimization

This study addresses two key issues, ie, the “cold-start problem” in transmission prediction of new or rare epidemics and the collaborative allocation of emergency medical resources considering multiple objectives. These two issues have not yet been well addressed in data-driven emergency medical resource allocation systems. A decision support prediction-then-optimization framework combing deep learning and optimization is developed to address these two issues. Two transfer learning based convolutional neural network models are built for epidemic transmission predictions in the initial and the subsequent outbreak regions using transfer learning to deal with the “cold-start problem”. A prediction-driven collaborative emergency medical resource allocation model is built to address the issue of collaborative decisions by simultaneously considering the inter- and intra-echelon resource flows in a multi-echelon system and considering the efficiency and fairness as the objective functions. A case study of the COVID-19 pandemic shows that combining transfer learning and convolutional neural networks can improve the performances of epidemic transmission predictions, and good predictions can improve both the efficiency and fairness of emergency medical resource allocation decisions. Moreover, the computational results show that the prediction errors are asymmetrically amplified in the optimization stage, and the shortage of the resource reserve quantity mediates the asymmetrical amplification effect. © Operational Research Society 2022.

PP-HumanSeg: Connectivity-Aware Portrait Segmentation with a Large-Scale Teleconferencing Video Dataset

As the COVID-19 pandemic rampages across the world, the demands of video conferencing surge. To this end, real-time portrait segmentation becomes a popular feature to replace backgrounds of conferencing participants. While feature-rich datasets, models and algorithms have been offered for segmentation that extract body postures from life scenes, portrait segmentation has yet not been well covered in a video conferencing context. To facilitate the progress in this field, we introduce an open-source solution named PP-HumanSeg. This work is the first to construct a large-scale video portrait dataset that contains 291 videos from 23 conference scenes with 14K fine-labeled frames and extensions to multi-camera teleconferencing. Furthermore, we propose a novel Self-supervised Connectivity-aware Learning (SCL) for semantic segmentation, which introduces a self-supervised connectivity-aware loss to improve the quality of segmentation results from the perspective of connectivity. And we propose an ultra-lightweight model with SCL for practical portrait segmentation, which achieves the best trade-off between IoU and the speed of inference. Extensive evaluations on our dataset demonstrate the superiority of SCL and our model.

Hospital admission, facility-based isolation, and social distancing: An SEIR model with constrained medical resources

COVID-19 is a highly contagious disease that has spread to most countries at unprecedented transmission speed. Medical resources and treatments provided by the healthcare system help reduce the mortality rate and spread of COVID-19 by isolating infectious individuals. We introduce a modified SEIR model that considers individuals access to limited medical resources to characterize the central role of medical resources during the pandemic. We discuss how the three hospital admission policies (hierarchy, mixed, and Fangcang healthcare system) affect the spread of the disease and the number of deaths and infections. We find that the Fangcang system results in the least number of infections, deaths, and occupied beds. When hospital capacity is relatively high or the transmission rate of the mildly infected patient is not ignorable, a mixed system can lead to fewer infections and deaths than a hierarchy system, but greater numbers of occupied beds. This occurs by preventing disease transmission to a great extent. The results are confirmed by our surveys with healthcare workers in major hospitals in Wuhan, China. We also investigate the performance of the three healthcare systems under a social distancing policy. We find that the Fangcang system results in the largest reduction in infections and deaths, especially even when the medical capacity is small. Moreover, we compare a one-time off policy with a bed trigger policy. We find that a one-time off policy could achieve the similar performance as bed trigger policy when it is initiated neither too early nor too late.

Drive High Power UVC-LED Wafer into Low-Cost 4-Inch Era: Effect of Strain Modulation

Ultraviolet-C light-emitting diodes (UVC-LEDs) have great application in pathogen inactivation under various kinds of situations, especially in the fight against COVID-19. Unfortunately, its epitaxial wafers are so far limited to a size of 2 inches, which greatly increases the cost of massive production. In this work, a 4-inch crack-free high-power UVC-LED wafer is reported. This achievement relies on a proposed strain-tailored strategy, where a 3D to 2D (3D-2D) transition layer is introduced during the homo-epitaxy of AlN on the high temperature annealed (HTA)-AlN template, which successfully drives the original compressive strain into a tensile one and thus solves the challenge of realizing a high-quality Al0.6Ga0.4N layer with a flat surface. This smooth Al0.6Ga0.4N layer is nearly pseudomorphically grown on the strain-tailored HTA-AlN template, leading to 4-inch UVC-LED wafers with outstanding performances. The strategy succeeds in compromising the bottlenecked contradictory in producing a large-sized UVC-LED wafer on pronounced crystalline AlN template: The compressive strain in HTA-AlN allows for a crack-free 4-inch wafer, but at the same time leads to a deterioration of the AlGaN morphology and crystal quality. The launch of 4-inch wafers makes the chip fabrication process of UVC-LEDs match the mature blue one, and will definitely speed up the universal application of UVC-LED in daily life.

The value of probabilistic forecasting in emergency medical resource planning under uncertainty

Purpose: Most epidemic transmission forecasting methods can only provide deterministic outputs. This study aims to show that probabilistic forecasting, in contrast, is suitable for stochastic demand modeling and emergency medical resource planning under uncertainty. Design/methodology/approach: Two probabilistic forecasting methods, i.e. quantile regression convolutional neural network and kernel density estimation, are combined to provide the conditional quantiles and conditional densities of infected populations. The value of probabilistic forecasting in improving decision performances and controlling decision risks is investigated by an empirical study on the emergency medical resource planning for the COVID-19 pandemic. Findings: The managerial implications obtained from the empirical results include (1) the optimization models using the conditional quantile or the point forecasting result obtain better results than those using the conditional density;(2) for sufficient resources, decision-makers' risk preferences can be incorporated to make tradeoffs between the possible surpluses and shortages of resources in the emergency medical resource planning at different quantile levels;and (3) for scarce resources, the differences in emergency medical resource planning at different quantile levels greatly decrease or disappear because of the existing of forecasting errors and supply quantity constraints. Originality/value: Very few studies concern probabilistic epidemic transmission forecasting methods, and this is the first attempt to incorporate deep learning methods into a two-phase framework for data-driven emergency medical resource planning under uncertainty. Moreover, the findings from the empirical results are valuable to select a suitable forecasting method and design an efficient emergency medical resource plan. © 2022, Emerald Publishing Limited.

A Framework for Situation-Aware Access Control in Federated Data-as-a-Service Systems Based on Query Rewriting

Organizations often need to share mission dependent data in a secure and flexible way. Examples include contact tracing for a contagious disease such as COVID-19, maritime search and rescue operations, or creating a collaborative bid for a contract. In such examples, the ability to access data may need to change dynamically, depending on the situation of a mission (e.g., whether a person tested positive for a disease, a ship is in distress, or a bid offer with given properties needs to be created). We present a novel framework to enable situation-aware access control in a federated Data-as-a-Service architecture by using semantic web technologies. Our framework allows distributed query rewriting and semantic reasoning that automatically adds situation based constraints to ensure that users can only see results that they are allowed to access. We have validated our framework by applying it to two dynamic use cases: maritime search and rescue operations and contact tracing for surveillance of a contagious disease. This paper details our implemented solution and experimental results of the two use cases. Our framework can be adopted by organizations that need to share sensitive data securely during dynamic, limited duration scenarios.

Blood molecular markers associated with COVID-19 immunopathology and multi-organ damage

COVID-19 is characterized by dysregulated immune responses, metabolic dysfunction and adverse effects on the function of multiple organs. To understand host responses to COVID-19 pathophysiology, we combined transcriptomics, proteomics, and metabolomics to identify molecular markers in peripheral blood and plasma samples of 66 COVID-19-infected patients experiencing a range of disease severities and 17 healthy controls. A large number of expressed genes, proteins, metabolites, and extracellular RNAs (exRNAs) exhibit strong associations with various clinical parameters. Multiple sets of tissue-specific proteins and exRNAs varied significantly in both mild and severe patients suggesting a potential impact on tissue function. Chronic activation of neutrophils, IFN-I signaling, and a high level of inflammatory cytokines were observed in patients with severe disease progression. In contrast, COVID-19-infected patients experiencing milder disease symptoms showed robust T-cell responses. Finally, we identified genes, proteins, and exRNAs as potential biomarkers that might assist in predicting the prognosis of SARS-CoV-2 infection. These data refine our understanding of the pathophysiology and clinical progress of COVID-19. SYNOPSIS image Proteomics, metabolomics and RNAseq data map immune responses in COVID-19 patients with different disease severity, revealing molecular makers associated with disease progression and alterations of tissue-specific proteins. A multi-omics profiling of the host response to SARS-CoV2 infection in 66 clinically diagnosed and laboratory confirmed COVID-19 patients and 17 uninfected controls. Significant correlations between multi-omics data and key clinical parameters. Alteration of tissue-specific proteins and exRNAs. Enhanced activation of immune responses is associated with COVID-19 pathogenesis. Biomarkers to predict COVID-19 clinical outcomes pending clinical validation as prospective marker.