Secure Convolutional Neural Network-based Internet-of-Healthcare Applications

Convolutional neural networks (CNNs) have gained popularity for Internet-of-Healthcare (IoH) applications such as medical diagnostics. However, new research shows that adversarial attacks with slight imperceptible changes can undermine deep neural network techniques in healthcare. This raises questions regarding the safety of deploying these IoH devices in clinical situations. In this paper, we review the techniques used in fighting against cyber-attacks. Then, we propose to study the robustness of some well-known CNN architectures’belonging to sequential, parallel, and residual families, such as LeNet5, MobileNetV1, VGG16, ResNet50, and InceptionV3 against fast gradient sign method (FGSM) and projected gradient descent (PGD) attacks, in the context of classification of chest radiographs (X-rays) based on the IoH application. Finally, we propose to improve the security of these CNN structures by studying standard and adversarial training. The results show that, among these models, smaller models with lower computational complexity are more secure against hostile threats than larger models that are frequently used in IoH applications. In contrast, we reveal that when these networks are learned adversarially, they can outperform standard trained networks. The experimental results demonstrate that the model performance breakpoint is represented by γ= 0.3 with a maximum loss of accuracy tolerated at 2%. Author

Post-Pandemic Follow-Up Audit of Security Checkpoints

This paper provides a follow-up audit of security checkpoints (or simply checkpoints) for mass transportation hubs such as airports and seaports aiming at the post-pandemic R&D adjustments. The goal of our study is to determine biometric-enabled resources of checkpoints for a counter-epidemic response. To achieve the follow-up audit goals, we embedded the checkpoint into the Emergency Management Cycle (EMC) –the core of any doctrine that challenges disaster. This embedding helps to identify the technology-societal gaps between contemporary and post-pandemic checkpoints. Our study advocates a conceptual exploration of the problem using EMC profiling and formulates new tasks for checkpoints based on the COVID-19 pandemic lessons learned. In order to increase practical value, we chose a case study of face biometrics for an experimental post-pandemic follow-up audit. Author

Nonlinear time-series forecasts for decision support: short-term demand for ICU beds in Santiago, Chile, during the 2021 COVID-19 pandemic.

In Chile, due to the explosive increase of new Coronavirus disease 2019 (COVID-19) cases during the first part of 2021, the ability of health services to accommodate new incoming cases was jeopardized. It has become necessary to be able to manage intensive care unit (ICU) capacity, and for this purpose, monitoring both the evolution of new cases and the demand for ICU beds has become urgent. This paper presents short-term forecast models for the number of new cases and the number of COVID-19 patients admitted to ICUs in the Metropolitan Region in Chile.

Machine Learning Models for Predicting Financially Vigilant Low-Income Households

The COVID-19 pandemic has adversely affected households’lives in terms of social and economic factors across the world. The Malaysian government has devised a number of stimulus packages to combat the pandemic’s effects. Stimulus packages would be insufficient to alleviate household financial burdens if they did not target those most affected by lockdowns. As a result, assessing household financial vigilance in the case of crisis like the COVID-19 pandemic is crucial. This study aimed to develop machine learning models for predicting and profiling financially vigilant households. The Special Survey on the Economic Effects of Covid-19 and Individual Round 1 provided secondary data for this study. As a research methodology, a cross-industry standard process for data mining is followed. Five machine learning algorithms were used to build predictive models. Among all, Gradient Boosted Tree was identified as the best predictive model based on F-score measure. The findings showed machine learning approach can provide a robust model to predict households’financial vigilances, and this information might be used to build appropriate and effective economic stimulus packages in the future. Researchers, academics and policymakers in the field of household finance can use these recommendations to help them leverage machine learning. Author

DeCoP:Deep Learning for COVID-19 Prediction of Survival

The current ongoing COVID-19 pandemic caused by the SARS-CoV-2 virus, has severely affected our daily life routines and behavior patterns. According to the World Health Organization, there have been 93 million confirmed cases with more than 1.99 million confirmed death around 235 Countries, areas or territories until 15 January 2021, 11:00 GMT+11. People who are affected with COVID-19 have different symptoms from people to people. When large amounts of patients are affected with COVID-19, it is important to quickly identify the health conditions of patients based on the basic information and symptoms of patients. Then the hospital can arrange reasonable medical resources for different patients. However, existing work has a low recall of 15.7% for survival predictions based on the basic information of patients (i.e., false positive rate (FPR) with 84.3%, FPR: actually survival but predicted as died). There is much room for improvement when using machine learning-based techniques for COVID-19 prediction. In this paper, we propose DeCoP to train a classifier to predict the survival of COVID-19 patients with high recall and F1 score. DeCoP is a deep learning (DL)-based scheme of Bidirectional Long Short-Term Memory (BiLSTM) along with Fuzzy-based Information Decomposition (FID) to predict the survival of patients. First of all, we apply FID oversampling to redistribute the training data of the Open COVID-19 Data Working Group. Then, we employ BiLSTM to learn the high-level feature representations from the redistributed dataset. After that, the high-level feature vector will be used to train the prediction model. Experimental results show that our proposed scheme achieves outstanding performances. Precisely, the improvement achieves about 19% and 18% in terms of recall and F1-measure. IEEE

Transportation and Location Planning During Epidemics/Pandemics: Emerging Problems and Solution Approaches

The sudden changes in human mobility, the immense increase in demand for logistics and delivery systems, governmental restrictions, and uncertainty of the spread dynamics have introduced several transportation and location-related decision problems during the COVID-19 pandemic. Hence, a variety of Operations Research (OR) tools and techniques have been applied to tackle these problems for mitigating the adverse effects of the spread. In this study, we first investigate the emerging decision problematics observed during epidemics/pandemics under four research clusters as: (i) effects of epidemics on transportation, (ii) effect of mobility on pandemic spread, (iii) logistics and delivery systems, and (iv) medical waste management and wastewater-based epidemiology. Next, we explore the OR tools implemented to solve the transportation and location-related decision problems in each cluster. IEEE

A machine learning approach to differentiate between COVID-19 and influenza infection using synthetic infection and immune response data.

Data analysis is widely used to generate new insights into human disease mechanisms and provide better treatment methods. In this work, we used the mechanistic models of viral infection to generate synthetic data of influenza and COVID-19 patients. We then developed and validated a supervised machine learning model that can distinguish between the two infections. Influenza and COVID-19 are contagious respiratory illnesses that are caused by different pathogenic viruses but appeared with similar initial presentations. While having the same primary signs COVID-19 can produce more severe symptoms, illnesses, and higher mortality. The predictive model performance was externally evaluated by the ROC AUC metric (area under the receiver operating characteristic curve) on 100 virtual patients from each cohort and was able to achieve at least AUC = 91% using our multiclass classifier. The current investigation highlighted the ability of machine learning models to accurately identify two different diseases based on major components of viral infection and immune response. The model predicted a dominant role for viral load and productively infected cells through the feature selection process.

Predictive Shoulder Kinematics of Rehabilitation Exercises through Immersive Virtual Reality

Objective: The adoption of telehealth rapidly accelerated due to the global COVID19 pandemic disrupting communities and in-person healthcare practices. While telehealth had initial benefits in enhancing accessibility for remote treatment, physical rehabilitation has been heavily limited due to the loss of hands-on evaluation tools. This paper presents an immersive virtual reality (iVR) pipeline for replicating physical therapy success metrics through applied machine learning of patient observation. Methods: We demonstrate a method of training gradient boosted decision-trees for kinematic estimation to replicate mobility and strength metrics with an off-the-shelf iVR system. During a two-month study, training data was collected while a group of users completed physical rehabilitation exercises in an iVR game. Utilizing this data, we trained on iVR based motion capture data and OpenSim biomechanical simulations. Results: Our final model indicates that upper-extremity kinematics from OpenSim can be accurately predicted using the HTC Vive head-mounted display system with a Mean Absolute Error less than 0.78 degrees for joint angles and less than 2.34 Nm for joint torques. Additionally, these predictions are viable for run-time estimation, with approximately a 0.74 ms rate of prediction during exercise sessions. Conclusion: These findings suggest that iVR paired with machine learning can serve as an effective medium for collecting evidence-based patient success metrics in telehealth. Significance: Our approach can help increase the accessibility of physical rehabilitation with off-the-shelf iVR head-mounted display systems by providing therapists with metrics needed for remote evaluation. Author

Advanced Muti-Mutation with Intervention Policies Pandemic Model

A pandemic is a threat to humanity with potentially millions of deaths worldwide. Epidemiological models can be used to better understand pandemic dynamics and assist policymakers in optimizing their Intervention Policies (IPs). Most existing epidemiological models assume, sometimes incorrectly, that a pandemic is caused by a single pathogen, ignoring pathogen mutations over time that result in different pathogen variants with different characteristics. In addition, the existing models do not incorporate the effect of IPs like vaccinations and lockdowns during the fitting phase. In this work, we introduce a new multi-mutation model called Suspected-Infected-Vaccinated-Recovered-reInfected (SIVRI). This model extends the SIRS model with adaptation to incorporate available knowledge related to the different pathogen mutations together with multiple IPs. In order to find the model parameters we propose a new fitting procedure that supports the complex social, epidemiological, and clinical dynamics that occur during a pandemic. We examine the suggested SIVRI model in comparison to the SIRS and XGboost models on the COVID-19 pandemic in Israel that includes four COVID-19 mutations, and the vaccination and lockdown IPs. We show that the proposed model can fit accurately to the historical data and outperform the existing models in predictions of basic reproduction number, mortality rate, and severely infected individuals rate. Author

Estimation of the Basic Reproduction Number for the COVID-19 Pandemic in Minnesota

This paper focuses on the dynamics of the COVID-19 pandemic and estimation of associated real-time variables characterizing disease spread. A nonlinear dynamic model is developed which enhances the traditional SEIR epidemic model to include additional variables of hospitalizations, ICU admissions, and deaths. A 6-month data set containing Minnesota data on infections, hospital-ICU admissions and deaths is used to find least-squares solutions to the parameters of the model. The model is found to fit the measured data accurately. Subsequently, a cascaded observer is developed to find real-time values of the infected population, the infection rate, and the basic reproduction number. The observer is found to yield good real-time estimates that match the least-squares parameters obtained from the complete data set. The importance of the work is that it enables real-time estimation of the basic reproduction number which is a key variable for controlling disease spread. Copyright © 2021 The Authors. This is an open access article under the CC BY-NC-ND license

A Review of Simulation and Application of Agent-Based Model Approaches

In the past, various traditional methods used experiments and statistical data to examine and solve the occurred problem and social-environmental issue. However, the traditional method is not suitable for expressing or solving the complex dynamics of human environmental crisis (such as the spread of diseases, natural disaster management, social problems, etc.). Therefore, the implementation of computational modelling methods such as Agent-Based Models (ABM) has become an effective technology for solving complex problems arising from the interpretation of human behaviour such as human society, environment, and biological systems. Overall, this article will outline the ABM model properties and its applications in the criminology, flood management, and the COVID-19 pandemic fields. In addition, this article will review the limitations that occurred to be overcome in the further development of the ABM model. © 2021 Institute of Physics Publishing. All rights reserved.

Modeling the Impact of Social Distancing on the COVID-19 Pandemic in a Low Transmission Setting

According to the World Health Organization and the CDC, social distancing is currently one of the most effective ways to slow the transmission of COVID-19. However, most existing epidemic models do not consider the impact of social distancing on the COVID-19 pandemic. In this article, we propose a new method to deterministic modeling of the effects of social distancing on the COVID-19 pandemic in a low transmission setting. Our model dynamic is expressed by a single predictive variable that satisfies an integro-differential equation. Once the dynamic variable is calculated, the process of agents from the normal state, infection state to rehabilitation state, or death state can be explored. Besides, an important parameter is added to the model to measure the impact of social distancing on epidemic transmission. We performed qualitative and quantitative experiments on various scenarios, and the results showed that 2 m is a safe social distancing on the COVID-19 pandemic in a low transmission setting. IEEE

The Basic Reproduction Number as a Loop Gain Matrix.

The COVID-19 pandemic and the disordered reactions of most governments made the importance of mathematical modelling and model-based predictions evident, even outside the scientific community. The basic reproduction number [Formula: see text] quickly entered the common jargon, as a concise but effective tool to communicate the spreading power of a disease and estimate, at least roughly, the possible outcomes of the epidemic. However, while [Formula: see text] is easily defined for simple models, its proper definition is more subtle for larger, state-of-the-art models. Here we show that it is nothing else than the spectral radius of the gain matrix of a linear system, and that this matrix generalizes [Formula: see text] in the computation of the vector-valued final epidemic size and epidemic threshold, in a large class of finite-dimensional SIR-like models.