Artificial Intelligence for Forecasting the Prevalence of COVID-19 Pandemic: An Overview

Spatio-temporal analysis of COVID-19 lockdown effect to survive in the US counties using ANN

This study aims to quantify the effectiveness of lockdown as a non-pharmacological solution for managing the COVID-19 pandemic. Daily COVID-19 death counts were collected for four states: California, Georgia, New Jersey, and South Carolina. The effectiveness of the lockdown was studied and the number of people saved during 7 days was evaluated. Five neural network models (MLP, FFNN, CFNN, ENN, and NARX) were implemented, and the results indicate that FFNN is the best prediction model. Based on this model, the total number of survivors over a 7-day period is 211, 270, 989, and 60 in California, Georgia, New Jersey, and South Carolina, respectively. The coefficients and weights of the FFNN for each state differ due to various factors, including socio-demographic conditions and the behavior of citizens towards lockdown laws. New Jersey and South Carolina have the most lockdowns and the least.

SVM-RFE enabled feature selection with DMN based centroid update model for incremental data clustering using COVID-19

This research introduces an efficacious model for incremental data clustering using Entropy weighted-Gradient Namib Beetle Mayfly Algorithm (NBMA). Here, feature selection is done based upon support vector machine recursive feature elimination (SVM-RFE), where the weight parameter is optimally fine-tuned using NBMA. After that, clustering is carried out utilizing entropy weighted power k-means clustering algorithm and weight is updated employing designed Gradient NBMA. Finally, incremental data clustering takes place in which centroid matching is carried out based on RV coefficient, whereas centroid is updated based on deep maxout network (DMN). Also, the result shows the better performance of the proposed method..

Forecasting the spread of COVID-19 based on policy, vaccination, and Omicron data

The COVID-19 pandemic caused by the novel SARS-COV-2 virus poses a great risk to the world. During the COVID-19 pandemic, observing and forecasting several important indicators of the epidemic (like new confirmed cases, new cases in intensive care unit, and new deaths for each day) helped prepare the appropriate response (e.g., creating additional intensive care unit beds, and implementing strict interventions). Various predictive models and predictor variables have been used to forecast these indicators. However, the impact of prediction models and predictor variables on forecasting performance has not been systematically well analyzed. Here, we compared the forecasting performance using a linear mixed model in terms of prediction models (mathematical, statistical, and AI/machine learning models) and predictor variables (vaccination rate, stringency index, and Omicron variant rate) for seven selected countries with the highest vaccination rates. We decided on our best models based on the Bayesian Information Criterion (BIC) and analyzed the significance of each predictor. Simple models were preferred. The selection of the best prediction models and the use of Omicron variant rate were considered essential in improving prediction accuracies. For the test data period before Omicron variant emergence, the selection of the best models was the most significant factor in improving prediction accuracy. For the test period after Omicron emergence, Omicron variant rate use was considered essential in deciding forecasting accuracy. For prediction models, ARIMA, lightGBM, and TSGLM generally performed well in both test periods. Linear mixed models with country as a random effect has proven that the choice of prediction models and the use of Omicron data was significant in determining forecasting accuracies for the highly vaccinated countries. Relatively simple models, fit with either prediction model or Omicron data, produced best results in enhancing forecasting accuracies with test data.

Unveiling the predictive power: a comprehensive study of machine learning model for anticipating chronic kidney disease

In today's modern era, chronic kidney disease stands as a significantly grave ailment that detrimentally impacts human life. This issue is progressively escalating in both developed and developing nations. Precise and timely identification of chronic kidney disease is imperative for the prevention and management of kidney failure. Historical methods of diagnosing chronic kidney disease have often been deemed unreliable on several fronts. To distinguish between healthy individuals and those afflicted by chronic kidney disease, dependable and effective non-invasive techniques such as machine learning models have been adopted. In our ongoing research, we employ various machine learning models, encompassing logistic regression, random forest, decision tree, k-nearest neighbor, and support vector machine utilizing four kernel functions (linear, Laplacian, Bessel, and radial basis kernels), to forecast chronic kidney disease. The dataset used constitutes records from a case-control study involving chronic kidney disease patients in district Buner, Khyber Pakhtunkhwa, Pakistan. For comparative evaluation of the models in terms of classification and accuracy, diverse performance metrics, including accuracy, Brier score, sensitivity, Youden's index, and F1 score, were computed.

Mechanisms of Piperacillin/Tazobactam Nephrotoxicity: Piperacillin/Tazobactam-Induced Direct Tubular Damage in Mice

Piperacillin/tazobactam (PT) is one of the most commonly prescribed antibiotics for critically ill patients in intensive care. PT has been reported to cause direct nephrotoxicity; however, the underlying mechanisms remain unknown. We investigated the mechanisms underlying PT nephrotoxicity using a mouse model. The kidneys and sera were collected 24 h after PT injection. Serum blood urea nitrogen (BUN), creatinine, neutrophil gelatinase-associated lipocalin (NGAL), and renal pathologies, including inflammation, oxidative stress, mitochondrial damage, and apoptosis, were examined. Serum BUN, creatinine, and NGAL levels significantly increased in PT-treated mice. We observed increased IGFBP7, KIM-1, and NGAL expression in kidney tubules. Markers of oxidative stress, including 8-OHdG and superoxide dismutase, also showed a significant increase, accompanied by mitochondrial damage and apoptosis. The decrease in the acyl-coA oxidase 2 and Bcl2/Bax ratio also supports that PT induces mitochondrial injury. An in vitro study using HK-2 cells also demonstrated mitochondrial membrane potential loss, indicating that PT induces mitochondrial damage. PT appears to exert direct nephrotoxicity, which is associated with oxidative stress and mitochondrial damage in the kidney tubular cells. Given that PT alone or in combination with vancomycin is the most commonly prescribed antibiotic in patients at high risk of acute kidney injury, caution should be exercised.

Data driven contagion risk management in low-income countries using machine learning applications with COVID-19 in South Asia

In the absence of real-time surveillance data, it is difficult to derive an early warning system and potential outbreak locations with the existing epidemiological models, especially in resource-constrained countries. We proposed a contagion risk index (CR-Index)-based on publicly available national statistics-founded on communicable disease spreadability vectors. Utilizing the daily COVID-19 data (positive cases and deaths) from 2020 to 2022, we developed country-specific and sub-national CR-Index for South Asia (India, Pakistan, and Bangladesh) and identified potential infection hotspots-aiding policymakers with efficient mitigation planning. Across the study period, the week-by-week and fixed-effects regression estimates demonstrate a strong correlation between the proposed CR-Index and sub-national (district-level) COVID-19 statistics. We validated the CR-Index using machine learning methods by evaluating the out-of-sample predictive performance. Machine learning driven validation showed that the CR-Index can correctly predict districts with high incidents of COVID-19 cases and deaths more than 85% of the time. This proposed CR-Index is a simple, replicable, and easily interpretable tool that can help low-income countries prioritize resource mobilization to contain the disease spread and associated crisis management with global relevance and applicability. This index can also help to contain future pandemics (and epidemics) and manage their far-reaching adverse consequences.

Artificial Intelligence Applied to Improve Scientific Reviews: The Antibacterial Activity of Cistus Plants as Proof of Concept

Reviews have traditionally been based on extensive searches of the available bibliography on the topic of interest. However, this approach is frequently influenced by the authors' background, leading to possible selection bias. Artificial intelligence applied to natural language processing (NLP) is a powerful tool that can be used for systematic reviews by speeding up the process and providing more objective results, but its use in scientific literature reviews is still scarce. This manuscript addresses this challenge by developing a reproducible tool that can be used to develop objective reviews on almost every topic. This tool has been used to review the antibacterial activity of Cistus genus plant extracts as proof of concept, providing a comprehensive and objective state of the art on this topic based on the analysis of 1601 research manuscripts and 136 patents. Data were processed using a publicly available Jupyter Notebook in Google Collaboratory here. NLP, when applied to the study of antibacterial activity of Cistus plants, is able to recover the main scientific manuscripts and patents related to the topic, avoiding any biases. The NLP-assisted literature review reveals that C. creticus and C. monspeliensis are the first and second most studied Cistus species respectively. Leaves and fruits are the most commonly used plant parts and methanol, followed by butanol and water, the most widely used solvents to prepare plant extracts. Furthermore, Staphylococcus. aureus followed by Bacillus. cereus are the most studied bacterial species, which are also the most susceptible bacteria in all studied assays. This new tool aims to change the actual paradigm of the review of scientific literature to make the process more efficient, reliable, and reproducible, according to Open Science standards.

Evolution of Antimicrobial Susceptibility to Penicillin in Invasive Strains of Streptococcus pneumoniae during 2007-2021 in Madrid, Spain

The use of pneumococcal conjugate vaccines has affected the epidemiology and distribution of Streptococcus pneumoniae serotypes causing Invasive Pneumococcal Disease (IPD). The aim of this study was to analyze the evolution of the phenotypical profiles of antimicrobial susceptibility to penicillin (PEN) in all IPD strains isolated in Madrid, Spain, during 2007-2021. In total, 7133 invasive clinical isolates were characterized between 2007 and 2021. Levels of PENR and PNSSDR were 2.0% and 24.2%, respectively. In addition, 94.4% of all the PENR belonged to four serotypes, including 11A (33.6%), 19A (30.8%), 14 (20.3%) and 9V (9.8%). All the strains of serotype 11A, which is a non-PCV13 serotype, were detected after the year 2011. Serotypes 6C, 15A, 23B, 24F, 35B, 19F, 16F, 6B, 23F, 24B, 24A, 15F and a limited number of strains of serogroups 16 and 24 (non-typed at serotype level) were associated with PNSSDR (p < 0.05). PNSSDR strains of non-PCV13 serotypes 11A, 24F, 23B, 24B, 23A and 16F were more frequent from 2014 to 2021. The changes in S. pneumoniae serotype distribution associated with the use of conjugate vaccines had caused in our region the emergence of non-PCV13 pneumococcal strains with different PENR or PNSSDR patterns. The emergence of serotype 11A resistant to penicillin as the most important non-PCV13 serotype is a worrisome event with marked relevance from the clinical and epidemiological perspective.

Wide-Scope Target and Suspect Screening of Antibiotics in Effluent Wastewater from Wastewater Treatment Plants in Europe

The occurrence of antibiotics in the environment could result in the development of antibiotic-resistant bacteria, which could result in a public health crisis. The occurrence of 676 antibiotics and the main transformation products (TPs) was investigated in the 48 wastewater treatment plants (WWTPs) from 11 countries (Germany, Romania, Serbia, Croatia, Slovenia, Hungary, Slovakia, Czechia, Austria, Cyprus, and Greece) by target and suspect screening. Target screening involved the investigation of antibiotics with reference standards (40 antibiotics). Suspect screening covered 676 antibiotics retrieved from the NORMAN Substance Database (antibiotic list on NORMAN network). Forty-seven antibiotics were detected in effluent wastewater samples: thirty-two by target screening and fifteen additional ones by suspect screening. An ecotoxicological risk assessment was performed based on occurrence data and predicted no effect concentration (PNEC), which involved the derivation of frequency of appearance (FoA), frequency of PNEC exceedance (FoE), and extent of PNEC exceedance (EoE). Azithromycin, erythromycin, clarithromycin, ofloxacin, and ciprofloxacin were prioritized as the calculated risk score was above 1. The median of antibiotics' load to freshwater ecosystems was 0.59 g/day/WWTP. The detection of antibiotics across countries indicates the presence of antibiotics in the ecosystems of Europe, which may trigger unwanted responses from the ecosystem, including antibiotic resistance.

Predicting Characteristics of Dissimilar Laser Welded Polymeric Joints Using a Multi-Layer Perceptrons Model Coupled with Archimedes Optimizer

This study investigates the application of a coupled multi-layer perceptrons (MLP) model with Archimedes optimizer (AO) to predict characteristics of dissimilar lap joints made of polymethyl methacrylate (PMMA) and polycarbonate (PC). The joints were welded using the laser transmission welding (LTW) technique equipped with a beam wobbling feature. The inputs of the models were laser power, welding speed, pulse frequency, wobble frequency, and wobble width; whereas, the outputs were seam width and shear strength of the joint. The Archimedes optimizer was employed to obtain the optimal internal parameters of the multi-layer perceptrons. In addition to the Archimedes optimizer, the conventional gradient descent technique, as well as the particle swarm optimizer (PSO), was employed as internal optimizers of the multi-layer perceptrons model. The prediction accuracy of the three models was compared using different error measures. The AO-MLP outperformed the other two models. The computed root mean square errors of the MLP, PSO-MLP, and AO-MLP models are (39.798, 19.909, and 2.283) and (0.153, 0.084, and 0.0321) for shear strength and seam width, respectively.

Predicting the Temperature Evolution during Nanomilling of Drug Suspensions via a Semi-Theoretical Lumped-Parameter Model

Although temperature can significantly affect the stability and degradation of drug nanosuspensions, temperature evolution during the production of drug nanoparticles via wet stirred media milling, also known as nanomilling, has not been studied extensively. This study aims to establish both descriptive and predictive capabilities of a semi-theoretical lumped parameter model (LPM) for temperature evolution. In the experiments, the mill was operated at various stirrer speeds, bead loadings, and bead sizes, while the temperature evolution at the mill outlet was recorded. The LPM was formulated and fitted to the experimental temperature profiles in the training runs, and its parameters, i.e., the apparent heat generation rate Q_gen and the apparent overall heat transfer coefficient times surface area UA, were estimated. For the test runs, these parameters were predicted as a function of the process parameters via a power law (PL) model and machine learning (ML) model. The LPM augmented with the PL and ML models was used to predict the temperature evolution in the test runs. The LPM predictions were also compared with those of an enthalpy balance model (EBM) developed recently. The LPM had a fitting capability with a root-mean-squared error (RMSE) lower than 0.9 °C, and a prediction capability, when augmented with the PL and ML models, with an RMSE lower than 4.1 and 2.1 °C, respectively. Overall, the LPM augmented with the PL model had both good descriptive and predictive capability, whereas the one with the ML model had a comparable predictive capability. Despite being simple, with two parameters and obviating the need for sophisticated numerical techniques for its solution, the semi-theoretical LPM generally predicts the temperature evolution similarly or slightly better than the EBM. Hence, this study has provided a validated, simple model for pharmaceutical engineers to simulate the temperature evolution during the nanomilling process, which will help to set proper process controls for thermally labile drugs.

Forecasting admissions in psychiatric hospitals before and during Covid-19: a retrospective study with routine data

The COVID-19 pandemic has strong effects on most health care systems. Forecasting of admissions can help for the efficient organisation of hospital care. We aimed to forecast the number of admissions to psychiatric hospitals before and during the COVID-19 pandemic and we compared the performance of machine learning models and time series models. This would eventually allow to support timely resource allocation for optimal treatment of patients. We used admission data from 9 psychiatric hospitals in Germany between 2017 and 2020. We compared machine learning models with time series models in weekly, monthly and yearly forecasting before and during the COVID-19 pandemic. A total of 90,686 admissions were analysed. The models explained up to 90% of variance in hospital admissions in 2019 and 75% in 2020 with the effects of the COVID-19 pandemic. The best models substantially outperformed a one-step seasonal naïve forecast (seasonal mean absolute scaled error (sMASE) 2019: 0.59, 2020: 0.76). The best model in 2019 was a machine learning model (elastic net, mean absolute error (MAE): 7.25). The best model in 2020 was a time series model (exponential smoothing state space model with Box-Cox transformation, ARMA errors and trend and seasonal components, MAE: 10.44). Models forecasting admissions one week in advance did not perform better than monthly and yearly models in 2019 but they did in 2020. The most important features for the machine learning models were calendrical variables. Model performance did not vary much between different modelling approaches before the COVID-19 pandemic and established forecasts were substantially better than one-step seasonal naïve forecasts. However, weekly time series models adjusted quicker to the COVID-19 related shock effects. In practice, multiple individual forecast horizons could be used simultaneously, such as a yearly model to achieve early forecasts for a long planning period and weekly models to adjust quicker to sudden changes.

Mathematical modeling and analysis of the SARS-Cov-2 disease with reinfection

The COVID-19 infection which is still infecting many individuals around the world and at the same time the recovered individuals after the recovery are infecting again. This reinfection of the individuals after the recovery may lead the disease to worse in the population with so many challenges to the health sectors. We study in the present work by formulating a mathematical model for SARS-CoV-2 with reinfection. We first briefly discuss the formulation of the model with the assumptions of reinfection, and then study the related qualitative properties of the model. We show that the reinfection model is stable locally asymptotically when R0<1. For R0≤1, we show that the model is globally asymptotically stable. Further, we consider the available data of coronavirus from Pakistan to estimate the parameters involved in the model. We show that the proposed model shows good fitting to the infected data. We compute the basic reproduction number with the estimated and fitted parameters numerical value is R0≈1.4962. Further, we simulate the model using realistic parameters and present the graphical results. We show that the infection can be minimized if the realistic parameters (that are sensitive to the basic reproduction number) are taken into account. Also, we observe the model prediction for the total infected cases in the future fifth layer of COVID-19 in Pakistan that may begin in the second week of February 2022.

Automated System for Identifying COVID-19 Infections in Computed Tomography Images Using Deep Learning Models

Coronavirus disease 2019 (COVID-19) is a novel disease that affects healthcare on a global scale and cannot be ignored because of its high fatality rate. Computed tomography (CT) images are presently being employed to assist doctors in detecting COVID-19 in its early stages. In several scenarios, a combination of epidemiological criteria (contact during the incubation period), the existence of clinical symptoms, laboratory tests (nucleic acid amplification tests), and clinical imaging-based tests are used to diagnose COVID-19. This method can miss patients and cause more complications. Deep learning is one of the techniques that has been proven to be prominent and reliable in several diagnostic domains involving medical imaging. This study utilizes a convolutional neural network (CNN), stacked autoencoder, and deep neural network to develop a COVID-19 diagnostic system. In this system, classification undergoes some modification before applying the three CT image techniques to determine normal and COVID-19 cases. A large-scale and challenging CT image dataset was used in the training process of the employed deep learning model and reporting their final performance. Experimental outcomes show that the highest accuracy rate was achieved using the CNN model with an accuracy of 88.30%, a sensitivity of 87.65%, and a specificity of 87.97%. Furthermore, the proposed system has outperformed the current existing state-of-the-art models in detecting the COVID-19 virus using CT images.

Application of Machine Learning Approaches to Predict the Strength Property of Geopolymer Concrete

Geopolymer concrete (GPC) based on fly ash (FA) is being studied as a possible alternative solution with a lower environmental impact than Portland cement mixtures. However, the accuracy of the strength prediction still needs to be improved. This study was based on the investigation of various types of machine learning (ML) approaches to predict the compressive strength (C-S) of GPC. The support vector machine (SVM), multilayer perceptron (MLP), and XGBoost (XGB) techniques have been employed to check the difference between the experimental and predicted results of the C-S for the GPC. The coefficient of determination (R²) was used to measure how accurate the results were, which usually ranged from 0 to 1. The results show that the XGB was a more accurate model, indicating an R² value of 0.98, as opposed to SVM (0.91) and MLP (0.88). The statistical checks and k-fold cross-validation (CV) also confirm the high precision level of the XGB model. The lesser values of the errors for the XGB approach, such as mean absolute error (MAE), mean square error (MSE), and root mean square error (RMSE), were noted as 1.49 MPa, 3.16 MPa, and 1.78 MPa, respectively. These lesser values of the errors also indicate the high precision of the XGB model. Moreover, the sensitivity analysis was also conducted to evaluate the parameter’s contribution towards the anticipation of C-S of GPC. The use of ML techniques for the prediction of material properties will not only reduce the effort of experimental work in the laboratory but also minimize the cast and time for the researchers.

Machine Learning to Rate and Predict the Efficiency of Waterflooding for Oil Production

Waterflooding is a widely used secondary oil recovery technique. The oil and gas industry uses a complex reservoir numerical simulation and reservoir engineering analysis to forecast production curves from waterflooding projects. The application of such standard methods at the stage of assessing the potential of a huge number of projects could be computationally inefficient and requires a lot of effort. This paper demonstrates the applicability of machine learning to rate the outcome of waterflooding applied to an oil reservoir. We also explore the relationship of project evaluations by operators at the final stages with several performance metrics for forecasting. Real data about several thousand waterflooding projects in Texas are used in the current study. We compare the ML models rankings of the waterflooding efficiency and the expert rankings. Linear regression models along with neural networks and gradient boosting on decision threes are considered. We show that machine learning models allow reducing computational complexity and can be useful for rating the reservoirs, with respect to the effectiveness of waterflooding.