Predicting Intensive Care Unit Admission Using Machine and Deep Learning: COVID-19 Case Study

The COVID-19 pandemic has affected the entire world by causing widespread panic and disrupting normal life. Since the outbreak began in December 2019, the virus has killed thousands of people and infected millions more. Hospitals are struggling to keep up with large patient flows. In some situations, hospitals are lacking enough beds and ventilators to accommodate all of their patients or are running low on supplies such as masks and gloves. Predicting intensive care unit (ICU) admission of patients with COVID-19 could help clinicians better allocate scarce ICU resources. In this study, many machine and deep learning algorithms are tested over predicting ICU admission of patients with COVID-19. Most of the algorithms we studied are extremely accurate toward this goal. With the convolutional neural network (CNN), we reach the highest results on our metrics (90.09% accuracy and 93.08% ROC-AUC), which demonstrates the usability of these learning models to identify patients who are likely to require ICU admission and assist hospitals in optimizing their resource management and allocation during the COVID-19 pandemic or others. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Research and Application of 5G Edge AI in Medical Industry

With the development of 5G and AI technology, the infectious virus detection framework system based on the combination of 5G MEC and medical sensors can effectively assist in the intelligent detection and control of influenza viruses such as COVID-19. Employing the edge computing and 5G+MEC model, the virus AI model is trained for the collected influenza virus data. Then the virus AI model can be used to evaluate the virus patients on the local edge computing service platform. Therefore, this paper introduces an algorithm and resource allocation, which uses 5G functions (especially, low latency, high bandwidth, wide connectivity, and other functions) to achieve local chest X-ray or CT scan images to detect COVID-19. Meanwhile, this paper also compares the computational efficiency of different algorithms in the 5G edge AI-based infectious virus detection framework, in this way to select the best algorithm and resource allocation. © 2022 IEEE.

Hospital Resource Management -A Survey

Healthcare is a limited resource that is constantly in high demand because everyone requires it. When demand exceeds supply, resources become relatively scarce, making the overall resource allocation in healthcare even more difficult, as we have seen at the time of COVID-19. Effective resource allocation faces obstacles such as a lack of trained human resources, inefficient resource use, a lack of focus on improvement, and inefficient resource reallocation. This paper will outline a study of the numerous approaches to resource allocation in healthcare, outlining the methods employed, the outcomes, and benefits and drawbacks of each approach. In order to address any kind of emergency situation that may arise in the future, it was our goal to pinpoint the research gap between the work that had already been done and the solution to this problem through the survey analysis. In order to boost hospital resource management, the paper identifies a variety of potential solutions which can be categorized further into subcategories which can be seen through different perspectives and a range of approaches that can be implemented during COVID-19 or in any other emergency condition. © 2023 IEEE.

HEALTH RESOURCE ALLOCATION METHOD AT CORDON SANITAIRE FOR PANDEMIC CONTROL AND PREVENTION

The outbreak of COVID-19 and its variants has profoundly disrupted our normal life. Many local authorities enforced cordon sanitaires for the protection of sensitive areas. Travelers can only cross the cordon after being tested. This paper aims to propose a method to determine the optimal deployment of cordon sanitaires in terms of minimum queueing delay time with available health testing resources. A sequential two-stage model is formulated where the first-stage model describes transportation system equilibrium to predict traffic ows. The second-stage model, a nonlinear integer programming, optimizes health resource allocation along the cordon sanitaire. This optimization aims to minimize the system's total delay time among all entry gates. Note that a stochastic queueing model is used to represent the queueing phenomenon at each entry link. A heuristic algorithm is designed to solve the proposed two-stage model where the Method of Successive Averages (MSA) is adopted for the first-stage model, and a genetic algorithm (GA) with elite strategy is adopted for the second-stage model. An experimental study is conducted to demonstrate the effectiveness of the proposed method and algorithm. The results show that these methods can find a good heuristic solution, and it is not cost-effective for authorities to keep adding health resources after reaching a certain limit. These methods are useful for policymakers to determine the optimal deployment of health resources at cordon sanitaires for pandemic control and prevention. © 2023.

A Methodology for Energy Usage Prediction in Long-Lasting Abnormal Events

Accurate energy consumption prediction is critical for proper resource allocation, meeting energy demand, and energy supply security. This work aims at developing a methodology for accurately modeling and predicting electricity consumption during abnormal long-lasting events, such as COVID-19 pandemic, which considerably affect consumption patterns in different types of premises. The proposed methodology involves three steps: (A) selects among multiple models the most accurate one in energy consumption prediction under normal conditions, (B) uses the selected model to analyze the impact of a specific abnormal event on energy consumption for various classes of premises, and (C) investigates which features contribute most to energy consumption prediction for abnormal conditions and which features can be added to improve such predictions.We use COVID-19 as a case study with datasets obtained from Fort Collins Utilities, which contain energy consumption data for residential and different sizes of commercial and industrial premises in the city of Fort Collins, Colorado, USA. We also use temperature records from NOAA and COVID-19 public orders from Larimer County.We validate the methodology by demonstrating that the methodology can help design a model suited for the pandemic situation using representative features, and as a result, accurately predict the energy consumption. Our results show that the MLP model selected by our methodology performs better than the other models even when they all use the COVID-related features. We also demonstrate that the methodology can help measure the impacts of the pandemic on the energy consumption. © 2022 IEEE.

Hierarchical Multi-agent Model for Reinforced Medical Resource Allocation with Imperfect Information

With the advent of the COVID-19 pandemic, the shortage in medical resources became increasingly more evident. Therefore, efficient strategies for medical resource allocation are urgently needed. However, conventional rule-based methods employed by public health experts have limited capability in dealing with the complex and dynamic pandemic-spreading situation. In addition, model-based optimization methods such as dynamic programming (DP) fail to work since we cannot obtain a precise model in real-world situations most of the time. Model-free reinforcement learning (RL) is a powerful tool for decision-making;however, three key challenges exist in solving this problem via RL: (1) complex situations and countless choices for decision-making in the real world;(2) imperfect information due to the latency of pandemic spreading;and (3) limitations on conducting experiments in the real world since we cannot set up pandemic outbreaks arbitrarily. In this article, we propose a hierarchical RL framework with several specially designed components. We design a decomposed action space with a corresponding training algorithm to deal with the countless choices, ensuring efficient and real-time strategies. We design a recurrent neural network-based framework to utilize the imperfect information obtained from the environment. We also design a multi-agent voting method, which modifies the decision-making process considering the randomness during model training and, thus, improves the performance. We build a pandemic-spreading simulator based on real-world data, serving as the experimental platform. We then conduct extensive experiments. The results show that our method outperforms all baselines, which reduces infections and deaths by 14.25% on average without the multi-agent voting method and up to 15.44% with it. © 2022 Association for Computing Machinery.

Resource distribution under spatiotemporal uncertainty of disease spread: Stochastic versus robust approaches

We consider the problem of optimizing locations of distribution centers (DCs) and plans for distributing resources such as test kits and vaccines, under spatiotemporal uncertainties of disease spread and demand for the resources. We aim to balance the operational cost (including costs of deploying facilities, shipping, and storage) and quality of service (reflected by demand coverage), while ensuring equity and fairness of resource distribution across multiple populations. We compare a sample-based stochastic programming (SP) approach with a distributionally robust optimization (DRO) approach using a moment-based ambiguity set. Numerical studies are conducted on instances of distributing COVID-19 vaccines in the United States and test kits, to compare SP and DRO models with a deterministic formulation using estimated demand and with the current resource distribution plans implemented in the US. We demonstrate the results over distinct phases of the pandemic to estimate the cost and speed of resource distribution depending on scale and coverage, and show the "demand-driven” properties of the SP and DRO solutions. Our results further indicate that if the worst-case unmet demand is prioritized, then the DRO approach is preferred despite of its higher overall cost. Nevertheless, the SP approach can provide an intermediate plan under budgetary restrictions without significant compromises in demand coverage. © 2022 Elsevier Ltd

A novel Energy Resources Allocation Management model for air pollution reduction.

Although air pollution has been reduced in various industrial and crowded cities during the COVID-19 pandemic, curbing the high concentration of the crisis of air pollution in the megacity of Tehran is still a challenging issue. Thus, identifying the major factors that play significant roles in increasing contaminant concentration is vital. This study aimed to propose a mathematical model to reduce air pollution in a way that does not require citizen participation, limitation on energy usage, alternative energies, any policies on fuel-burn style, extra cost, or time to ensure that consumers have access to energy adequately. In this study, we proposed a novel framework, denoted as the Energy Resources Allocation Management (ERAM) model, to reduce air pollution. The ERAM is designed to optimize the allocation of various energies to the recipients. To do so, the ERAM model is simulated based on the magnitude of fuel demand consumption, the rate of air pollution emission generated by each energy per unit per consumer, and the air pollution contribution produced by each user. To evaluate the reflectiveness and illustrate the feasibility of the model, a real-world case study, i.e., Tehran, was employed. The air pollution emission factors in Tehran territory were identified by considering both mobile sources, e.g., motorcycles, cars, and heavy-duty vehicles, and stationary sources, e.g., energy conversion stations, industries, and household and commercial sectors, which are the main contributors to particulate matter and nitrogen dioxide. An elaborate view of the results indicates that the ERAM model on fuel distribution could remarkably reduce Tehran's air pollution concentration by up to 14%.

Modeling Access Differences to Reduce Disparity in Resource Allocation

Motivated by COVID-19 vaccine allocation, where vulnerable subpopulations are simultaneously more impacted in terms of health and more disadvantaged in terms of access to the vaccine, we formalize and study the problem of resource allocation when there are inherent access differences that correlate with advantage and disadvantage. We identify reducing resource disparity as a key goal in this context and show its role as a proxy to more nuanced downstream impacts. We develop a concrete access model that helps quantify how a given allocation translates to resource flow for the advantaged vs. the disadvantaged, based on the access gap between them. We then provide a methodology for access-aware allocation. Intuitively, the resulting allocation leverages more vaccines in locations with higher vulnerable populations to mitigate the access gap and reduce overall disparity. Surprisingly, knowledge of the access gap is often not needed to perform access-aware allocation. To support this formalism, we provide empirical evidence for our access model and show that access-aware allocation can significantly reduce resource disparity and thus improve downstream outcomes. We demonstrate this at various scales, including at county, state, national, and global levels. © 2022 Owner/Author.

Avoiding Crisis Conditions in the Healthcare Infrastructure: 2 Case Studies in Statewide Collaboration.

In fall 2020, COVID-19 infections accelerated across the United States. For many states, a surge in COVID-19 cases meant planning for the allocation of scarce resources. Crisis standards of care planning focuses on maintaining high-quality clinical care amid extreme operating conditions. One of the primary goals of crisis standards of care planning is to use all preventive measures available to avoid reaching crisis conditions and the complex triage decisionmaking involved therein. Strategies to stay out of crisis must respond to the actual experience of people on the frontlines, or the "ground truth," to ensure efforts to increase critical care bed numbers and augment staff, equipment, supplies, and medications to provide an effective response to a public health emergency. Successful management of a surge event where healthcare needs exceed capacity requires coordinated strategies for scarce resource allocation. In this article, we examine the ground truth challenges encountered in response efforts during the fall surge of 2020 for 2 states-Nebraska and California-and the strategies each state used to enable healthcare facilities to stay out of crisis standards of care. Through these 2 cases, we identify key tools deployed to reduce surge and barriers to coordinated statewide support of the healthcare infrastructure. Finally, we offer considerations for operationalizing key tools to alleviate surge and recommendations for stronger statewide coordination in future public health emergencies.

A Dynamic Resource Allocation Framework for Synchronizing Metaverse with IoT Service and Data

Spurred by the severe restrictions on mobility due to the COVID-19 pandemic, there is currently intense interest in developing the Metaverse, to offer virtual services/business online. A key enabler of such virtual service is the digital twin, i.e., a digital replication of real-world entities in the Metaverse, e.g., city twin, avatars, etc. The real-world data collected by IoT devices and sensors are key for synchronizing the two worlds. In this paper, we consider the scenario in which a group of IoT devices are employed by the Metaverse platform to collect such data on behalf of virtual service providers (VSPs). Device owners, who are self-interested, dynamically select a VSP to maximize rewards. We adopt hybrid evolutionary dynamics, in which heterogeneous device owner populations can employ different revision protocols to update their strategies. Extensive simulations demonstrate that a hybrid protocol can lead to evolutionary stable states. © 2022 IEEE.

COV-HM: Prediction of COVID-19 Patient's Hospitalization Period for Hospital Management Using SMOTE and Machine Learning Techniques

COVID-19 imposes burdens on hospitals. Evidence-based management and optimum resource allocation are essential. Understanding the time frame of support needs for COVID-19 patients staying in hospitals is vital for planning hospital resource allocation, especially in resource-constrained settings. Machine learning methods are being utilized in the approximation of the length of stay of a patient in the hospital. Four machine learning classifiers were used in this study to estimate the duration of hospitalization for patients in 11 different classes. Due to the dataset's imbalance, SMOTE was applied to eliminate the problem. The prediction accuracy of the K-Nearest Neighbors, Random Forest, Decision Tree, and Gradient Boosting classifiers was 73%, 69%, 58%, and 57%. The feature importance scores assist in the identification of vital features while building machine learning models. This research will assist responsible authorities in maintaining hospital services depending on the length of a patient's stay. © 2022 ACM.

Virtual Machine allocation in multiple Data Centers using Throttled Load Balancing to improve the performance in Cloud

Today, Cloud Computing is a distributed system environment. These days the services are available pay as you go model. Cloud users are paying as per their services in the cloud environment. The services available to the Cloud users are Infrastructure as a service, platform as a service, software as a service and security as a service. Nowadays, most users are migrating to cloud platforms. In Covid-19 pandemic situation, most large and small scale organizations operating their business using cloud platforms. On the other end due to industrial automation, the companies switched their operations to a cloud environments. Due to the rapid business migration, the demand for cloud computing increased. With the increase of demand in the cloud, the service providers are satisfied. On the other end, a challenging issue is resource allocation. The best resource allocation strategy will provide quick services to the cloud users and minimum cost to the cloud providers. In this paper, we will discuss, resource allocation procedure, the throttled load balancing algorithm and the results are compared with other resource optimization techniques. © 2022 IEEE.

Adaptive resources allocation CUSUM for binomial count data monitoring with application to COVID-19 hotspot detection

In this paper, we present an efficient statistical method (denoted as 'Adaptive Resources Allocation CUSUM') to robustly and efficiently detect the hotspot with limited sampling resources. Our main idea is to combine the multi-arm bandit (MAB) and change-point detection methods to balance the exploration and exploitation of resource allocation for hotspot detection. Further, a Bayesian weighted update is used to update the posterior distribution of the infection rate. Then, the upper confidence bound (UCB) is used for resource allocation and planning. Finally, CUSUM monitoring statistics to detect the change point as well as the change location. For performance evaluation, we compare the performance of the proposed method with several benchmark methods in the literature and showed the proposed algorithm is able to achieve a lower detection delay and higher detection precision. Finally, this method is applied to hotspot detection in a real case study of county-level daily positive COVID-19 cases in Washington State WA) and demonstrates the effectiveness with very limited distributed samples.

Facing the Impact of the COVID-19 Pandemic: How Can We Allocate Outpatient Doctor Resources More Effectively?

The COVID-19 pandemic caused significant damage to global healthcare systems. Previous studies regarding COVID-19's impact on outpatient numbers focused only on a specific department, lacking research data for multiple departments in general hospitals. We assessed differences in COVID-19's impact on outpatient numbers for different departments to help hospital managers allocate outpatient doctor resources more effectively during the pandemic. We compared the outpatient numbers of 24 departments in a general hospital in Beijing in 2019 and 2020. We also examined an indicator not mentioned in previous studies, monthly departmental patient reservation rates. The results show that, compared with 2019, 2020 outpatient numbers decreased overall by 33.36%. Ten departments' outpatient numbers decreased >33.36%; however, outpatient numbers increased in two departments. In 2020, the overall patient reservation rate in 24 departments was 82.22% of the 2019 reservation rate; the rates in 14 departments were <82.22%. Moreover, patient reservation rates varied across different months. Our research shows that COVID-19's impact on different departments also varied. Additionally, our research suggests that well-known departments will be less affected by COVID-19, as will departments related to tumor treatment, where there may also be an increase in patient numbers. Patient reservation rates are an indicator worthy of attention. We suggest that hospital managers classify departments according to changes in outpatient numbers and patient reservation rates and adopt accurate, dynamic, and humanized management strategies to allocate outpatient doctor resources.

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.

Research on the Equity and Influencing Factors of Medical and Health Resources Allocation in the Context of COVID-19: A Case of Taiyuan, China.

COVID-19 has killed millions of people worldwide. As a result, medical and health resources continue to be strained, posing a great threat to people's safety and economic and social development. This paper built the index system of influencing factors of medical and health resources containing the economy, population and society, and then classified Taiyuan into three types of regions by cluster analysis. The Gini coefficient, Theil index and agglomeration degree were then used to analyze the spatial distribution of medical and health resources allocation, and its influencing factors were studied by grey relational analysis. It was found that the population allocation of medical and health resources in Taiyuan was better than area allocation. Population has the greatest influence on the allocation of medical and health resources, followed by society and the economy. The more developed the regional economy, the more diversified the main influencing factors, and the more adjustment and control choices of medical and health resources allocation. Suggestions for optimal allocation were put forward in order to fully utilize the limited medical and health resources, effectively respond to the epidemic needs, promote the sustainable development of resources, protect the health of residents, and improve social benefits.

Optimal Resource Allocation during Epidemics using Reinforcement Learning

Epidemics can prove to be disastrous, which has been further emphasized by the recent COVID-19 pandemic, and several countries like India lack sufficient resources to meet the population's needs. It is therefore important that the limited testing and protective resources are utilized such that the disease spread is minimized and their reach to the most vulnerable demographic is maximized. This paper studies the scope of intelligent agents in aiding authorities with such policy-making decisions. This is done by exploring the performance of various action selection methods on custom environments dealing with socio-economic groups and Indian states. Experiments using multi-armed bandit techniques provide greater insight into administrative decisions surrounding resource allocation and their future potential for greater use in similar scenarios. © 2022 IEEE.

The Balanced Allocation of Medical and Health Resources in Urban Areas of China from the Perspective of Sustainable Development: A Case Study of Nanjing

The balanced allocation of medical and health resources is an important basis for the sustainable development of health undertakings. In recent years, China has made remarkable achievements in the medical and health services, but there is still a phenomenon of unbalanced allocation of medical and health resources among different regions, which has become an urgent problem to be solved in deepening the reform of the medical and health system during the 14th Five-Year Plan period. From the perspective of people’s needs for health, this study analyzed the equity and efficiency of urban medical and health resources allocation in China by using the Theil index method and DEA method. Meanwhile, the authors used the coupling coordination degree model to construct a balanced development model with equity and efficiency as subsystems, taking the city of Nanjing as an example to analyze its balanced allocation of medical and health resources from 2008 to 2019. In general, taking Nanjing as an example, it shows that the balanced allocation of medical and health resources in Chinese cities is good, but in geographical dimension, the level of balanced allocation is low, and there are still significant differences in the equity and efficiency of allocation among regions. In the future, the government can strengthen the rationality of regional planning, appropriately increasing health investment and medical supply, considering both equity and efficiency to further realize the balanced allocation of medical and health resources and improve the sustainability of urban medical service system. The main contribution of this paper lies in that, from the perspective of sustainable development, the evaluation system is integrated to measure the equity and efficiency respectively, and the balanced development model is used to investigate the allocation of urban medical and health resources. The research results can provide reference for optimizing resources allocation and promoting the sustainable development of medical and health undertakings.

Predicting the Severity of COVID-19 Respiratory Illness with Deep Learning

Patient care in emergency rooms can utilize urgency labeling to facilitate resource allocation. With COVID-19 care, one of the most important indicators of care urgency is the severity of respiratory illness. We present an early analysis of 5,584 patient records, of whom 5,371 (96.2%) have returned a positive COVID-19 test, to understand how well we can predict the severity of a respiratory illness given other features describing a patient using Deep Learning methods. The goal of our work is to illustrate the connection of our COVID-19 patient dataset with Deep Learning techniques, setting the stage for future work. The features in our dataset include when COVID-19 symptoms began, age, height, weight, demographics, and pre-existing conditions, to give a quick preview. We report train-test performance of a Deep Multi-Layer Perceptron (MLP) to predict the severity of respiratory analysis on a one-hot encoded scale of 5 labels. This 5-level scale is a truncation of our available labels, which we plan to extend and include in future work. We utilize a high-level of Dropout in order to avoid overfitting with our Deep Learning model. Further, we particularly study the impact of class imbalance on this dataset (Johnson and Khoshgoftaar 2019). We find that Random Oversampling (ROS) is an effective solution for decreasing minority class false negatives, as well as increasing overall accuracy. Readers will understand the performance of Deep Learning, with Dropout and ROS, to predict the severity of a COVID-19 pa-tient’s respiratory illness in which patients are described with Tabular Electronic Health Records (EHR). © 2022 by the authors. All rights reserved.