COVID-19 prediction models: a systematic literature review

Climate change, its impact on emerging infectious diseases and new technologies to combat the challenge

ABSTRACTImproved sanitation, increased access to health care, and advances in preventive and clinical medicine have reduced the mortality and morbidity rates of several infectious diseases. However, recent outbreaks of several emerging infectious diseases (EIDs) have caused substantial mortality and morbidity, and the frequency of these outbreaks is likely to increase due to pathogen, environmental, and population effects driven by climate change. Extreme or persistent changes in temperature, precipitation, humidity, and air pollution associated with climate change can, for example, expand the size of EID reservoirs, increase host-pathogen and cross-species host contacts to promote transmission or spillover events, and degrade the overall health of susceptible host populations leading to new EID outbreaks. It is therefore vital to establish global strategies to track and model potential responses of candidate EIDs to project their future behaviour and guide research efforts on early detection and diagnosis technologies and vaccine development efforts for these targets. Multi-disciplinary collaborations are demanding to develop effective inter-continental surveillance and modelling platforms that employ artificial intelligence to mitigate climate change effects on EID outbreaks. In this review, we discuss how climate change has increased the risk of EIDs and describe novel approaches to improve surveillance of emerging pathogens that pose the risk for EID outbreaks, new and existing measures that could be used to contain or reduce the risk of future EID outbreaks, and new methods to improve EID tracking during further outbreaks to limit disease transmission.

Leveraging advances in data-driven deep learning methods for hybrid epidemic modeling

Mathematical modeling of epidemic dynamics is crucial to understand its underlying mechanisms, quantify important parameters, and make predictions that facilitate more informed decision-making. There are three major types of models: mechanistic models including the SEIR-type paradigm, alternative data-driven (DD) approaches, and hybrid models that combine mechanistic models with DD approaches. In this paper, we summarize our work in the COVID-19 Scenario Modeling Hub (SMH) for more than 12 rounds since early 2021 for informed decision support. We emphasize the importance of deep learning techniques for epidemic modeling via a flexible DD framework that substantially complements the mechanistic paradigm to evaluate various future epidemic scenarios. We start with a traditional curve-fitting approach to model cumulative COVID-19 based on the underlying SEIR-type mechanisms. Hospitalizations and deaths are modeled as binomial processes of cases and hospitalization, respectively. We further formulate two types of deep learning models based on multivariate long short term memory (LSTM) to address the challenges of more traditional DD models. The first LSTM is structurally similar to the curve fitting approach and assumes that hospitalizations and deaths are binomial processes of cases. Instead of using a predefined exponential curve, LSTM relies on the underlying data to identify the most appropriate functions, and is capable of capturing both long-term and short-term epidemic behaviors. We then relax the assumption of dependent inputs among cases, hospitalizations, and death. Another type of LSTM that handles all input time series as parallel signals, the independent multivariate LSTM, is developed. Independent multivariate LSTM can incorporate a wide range of data sources beyond traditional case-based epidemiological surveillance. The DD framework unleashes its potential in big data era with previously neglected heterogeneous surveillance data sources, such as syndromic, environment, genomic, serologic, infoveillance, and mobility data. DD approaches, especially LSTM, complement and integrate with the mechanistic modeling paradigm, provide a feasible alternative approach to model today's complex socio-epidemiological systems, and further leverage our ability to explore different scenarios for more informed decision-making during health emergencies.

A comprehensive benchmark for COVID-19 predictive modeling using electronic health records in intensive care

The COVID-19 pandemic highlighted the need for predictive deep-learning models in health care. However, practical prediction task design, fair comparison, and model selection for clinical applications remain a challenge. To address this, we introduce and evaluate two new prediction tasks-outcome-specific length-of-stay and early-mortality prediction for COVID-19 patients in intensive care-which better reflect clinical realities. We developed evaluation metrics, model adaptation designs, and open-source data preprocessing pipelines for these tasks while also evaluating 18 predictive models, including clinical scoring methods and traditional machine-learning, basic deep-learning, and advanced deep-learning models, tailored for electronic health record (EHR) data. Benchmarking results from two real-world COVID-19 EHR datasets are provided, and all results and trained models have been released on an online platform for use by clinicians and researchers. Our efforts contribute to the advancement of deep-learning and machine-learning research in pandemic predictive modeling.

AI-powered COVID-19 forecasting: a comprehensive comparison of advanced deep learning methods

BACKGROUND

The coronavirus disease 2019 (COVID-19) pandemic continues to pose significant challenges to the public health sector, including that of the United Arab Emirates (UAE). The objective of this study was to assess the efficiency and accuracy of various deep-learning models in forecasting COVID-19 cases within the UAE, thereby aiding the nation's public health authorities in informed decision-making.

METHODS

This study utilized a comprehensive dataset encompassing confirmed COVID-19 cases, demographic statistics, and socioeconomic indicators. Several advanced deep learning models, including long short-term memory (LSTM), bidirectional LSTM, convolutional neural network (CNN), CNN-LSTM, multilayer perceptron, and recurrent neural network (RNN) models, were trained and evaluated. Bayesian optimization was also implemented to fine-tune these models.

RESULTS

The evaluation framework revealed that each model exhibited different levels of predictive accuracy and precision. Specifically, the RNN model outperformed the other architectures even without optimization. Comprehensive predictive and perspective analytics were conducted to scrutinize the COVID-19 dataset.

CONCLUSION

This study transcends academic boundaries by offering critical insights that enable public health authorities in the UAE to deploy targeted data-driven interventions. The RNN model, which was identified as the most reliable and accurate for this specific context, can significantly influence public health decisions. Moreover, the broader implications of this research validate the capability of deep learning techniques in handling complex datasets, thus offering the transformative potential for predictive accuracy in the public health and healthcare sectors.

The reporting completeness and transparency of systematic reviews of prognostic prediction models for COVID-19 was poor: a methodological overview of systematic reviews

OBJECTIVES

To conduct a methodological overview of reviews to evaluate the reporting completeness and transparency of systematic reviews (SRs) of prognostic prediction models (PPMs) for COVID-19.

STUDY DESIGN AND SETTING

MEDLINE, Scopus, Cochrane Database of Systematic Reviews, and Epistemonikos (epistemonikos.org) were searched for SRs of PPMs for COVID-19 until December 31, 2022. The risk of bias in systematic reviews tool was used to assess the risk of bias. The protocol for this overview was uploaded in the Open Science Framework (https://osf.io/7y94c).

RESULTS

Ten SRs were retrieved; none of them synthesized the results in a meta-analysis. For most of the studies, there was absence of a predefined protocol and missing information on study selection, data collection process, and reporting of primary studies and models included, while only one SR had its data publicly available. In addition, for the majority of the SRs, the overall risk of bias was judged as being high. The overall corrected covered area was 6.3% showing a small amount of overlapping among the SRs.

CONCLUSION

The reporting completeness and transparency of SRs of PPMs for COVID-19 was poor. Guidance is urgently required, with increased awareness and education of minimum reporting standards and quality criteria. Specific focus is needed in predefined protocol, information on study selection and data collection process, and in the reporting of findings to improve the quality of SRs of PPMs for COVID-19.

The Telemedicine Demand Index and its Utility in Managing COVID-19 Case Surges

Introduction: Telemedicine was an integral component in Singapore's COVID-19 management strategy, having been deployed at a national level in a centrally-administered program whereby patients at higher risk of developing severe COVID-19 disease were proactively assigned tele-consultations, whereas those at lower risk and seen by primary care physicians could request ad hoc tele-consultations. To better plan for fluctuations in telemedicine demand during the pandemic, the Telemedicine Demand Index (TDI) was developed. Methods: Three main factors influencing telemedicine demand were considered-characteristics of the Variant of Concern, prevailing health care policies, and the population's healthcare-seeking behaviour-from which 11 coefficients were derived for the TDI formula. The number of tele-consultations demanded is the product of the TDI and the total number of new COVID-19 cases for a given period. Results: Real-world data from January 31 to March 27, 2022 were compared with TDI estimates. A total of 148,485 tele-consultations were conducted against a backdrop of 723,675 new COVID-19 cases for the period. The TDI overestimated demand by an average 11.4%. Data from March 28 to May 1, 2022 were then used to derive new TDI values and applied to a 3-week period starting May 9, 2022, following a policy change. A total of 5,560 tele-consultations were conducted against a backdrop of 77,998 new COVID-19 cases. The TDI underestimated demand by an average of 7.2%. Conclusion: The TDI shows initial promise for quickly estimating telemedicine demand at a population level. By leveraging historical data and applying some informed assumptions, it allows for the estimation of current capabilities and future requirements. There remains scope for more research to refine the TDI's constituent components, as well as its applicability in different population contexts.

Vaccination compartmental epidemiological models for the delta and omicron SARS-CoV-2 variants

We explore the inclusion of vaccination in compartmental epidemiological models concerning the delta and omicron variants of the SARS-CoV-2 virus that caused the COVID-19 pandemic. We expand on our earlier compartmental-model work by incorporating vaccinated populations. We present two classes of models that differ depending on the immunological properties of the variant. The first one is for the delta variant, where we do not follow the dynamics of the vaccinated individuals since infections of vaccinated individuals were rare. The second one for the far more contagious omicron variant incorporates the evolution of the infections within the vaccinated cohort. We explore comparisons with available data involving two possible classes of counts, fatalities and hospitalizations. We present our results for two regions, Andalusia and Switzerland (including the Principality of Liechtenstein), where the necessary data are available. In the majority of the considered cases, the models are found to yield good agreement with the data and have a reasonable predictive capability beyond their training window, rendering them potentially useful tools for the interpretation of the COVID-19 and further pandemic waves, and for the design of intervention strategies during these waves.

Development and validation of COEWS (COVID-19 Early Warning Score) for hospitalized COVID-19 with laboratory features: A multicontinental retrospective study

Background

The emergence of new SARS-CoV-2 variants with significant immune-evasiveness, the relaxation of measures for reducing the number of infections, the waning of immune protection (particularly in high-risk population groups), and the low uptake of new vaccine boosters, forecast new waves of hospitalizations and admission to intensive care units. There is an urgent need for easily implementable and clinically effective Early Warning Scores (EWSs) that can predict the risk of complications within the next 24-48 hr. Although EWSs have been used in the evaluation of COVID-19 patients, there are several clinical limitations to their use. Moreover, no models have been tested on geographically distinct populations or population groups with varying levels of immune protection.

Methods

We developed and validated COVID-19 Early Warning Score (COEWS), an EWS that is automatically calculated solely from laboratory parameters that are widely available and affordable. We benchmarked COEWS against the widely used NEWS2. We also evaluated the predictive performance of vaccinated and unvaccinated patients.

Results

The variables of the COEWS predictive model were selected based on their predictive coefficients and on the wide availability of these laboratory variables. The final model included complete blood count, blood glucose, and oxygen saturation features. To make COEWS more actionable in real clinical situations, we transformed the predictive coefficients of the COEWS model into individual scores for each selected feature. The global score serves as an easy-to-calculate measure indicating the risk of a patient developing the combined outcome of mechanical ventilation or death within the next 48 hr.The discrimination in the external validation cohort was 0.743 (95% confidence interval [CI]: 0.703-0.784) for the COEWS score performed with coefficients and 0.700 (95% CI: 0.654-0.745) for the COEWS performed with scores. The area under the receiver operating characteristic curve (AUROC) was similar in vaccinated and unvaccinated patients. Additionally, we observed that the AUROC of the NEWS2 was 0.677 (95% CI: 0.601-0.752) in vaccinated patients and 0.648 (95% CI: 0.608-0.689) in unvaccinated patients.

Conclusions

The COEWS score predicts death or MV within the next 48 hr based on routine and widely available laboratory measurements. The extensive external validation, its high performance, its ease of use, and its positive benchmark in comparison with the widely used NEWS2 position COEWS as a new reference tool for assisting clinical decisions and improving patient care in the upcoming pandemic waves.

Funding

University of Vienna.

Harnessing the power of AI: Advanced deep learning models optimization for accurate SARS-CoV-2 forecasting

The pandemic has significantly affected many countries including the USA, UK, Asia, the Middle East and Africa region, and many other countries. Similarly, it has substantially affected Malaysia, making it crucial to develop efficient and precise forecasting tools for guiding public health policies and approaches. Our study is based on advanced deep-learning models to predict the SARS-CoV-2 cases. We evaluate the performance of Long Short-Term Memory (LSTM), Bi-directional LSTM, Convolutional Neural Networks (CNN), CNN-LSTM, Multilayer Perceptron, Gated Recurrent Unit (GRU), and Recurrent Neural Networks (RNN). We trained these models and assessed them using a detailed dataset of confirmed cases, demographic data, and pertinent socio-economic factors. Our research aims to determine the most reliable and accurate model for forecasting SARS-CoV-2 cases in the region. We were able to test and optimize deep learning models to predict cases, with each model displaying diverse levels of accuracy and precision. A comprehensive evaluation of the models' performance discloses the most appropriate architecture for Malaysia's specific situation. This study supports ongoing efforts to combat the pandemic by offering valuable insights into the application of sophisticated deep-learning models for precise and timely SARS-CoV-2 case predictions. The findings hold considerable implications for public health decision-making, empowering authorities to create targeted and data-driven interventions to limit the virus's spread and minimize its effects on Malaysia's population.

Is It Possible to Predict COVID-19? Stochastic System Dynamic Model of Infection Spread in Kazakhstan

BACKGROUND

Since the start of the COVID-19 pandemic, scientists have begun to actively use models to determine the epidemiological characteristics of the pathogen. The transmission rate, recovery rate and loss of immunity to the COVID-19 virus change over time and depend on many factors, such as the seasonality of pneumonia, mobility, testing frequency, the use of masks, the weather, social behavior, stress, public health measures, etc. Therefore, the aim of our study was to predict COVID-19 using a stochastic model based on the system dynamics approach.

METHOD

We developed a modified SIR model in AnyLogic software. The key stochastic component of the model is the transmission rate, which we consider as an implementation of Gaussian random walks with unknown variance, which was learned from real data.

RESULTS

The real data of total cases turned out to be outside the predicted minimum-maximum interval. The minimum predicted values of total cases were closest to the real data. Thus, the stochastic model we propose gives satisfactory results for predicting COVID-19 from 25 to 100 days. The information we currently have about this infection does not allow us to make predictions with high accuracy in the medium and long term.

CONCLUSIONS

In our opinion, the problem of the long-term forecasting of COVID-19 is associated with the absence of any educated guess regarding the dynamics of β(t) in the future. The proposed model requires improvement with the elimination of limitations and the inclusion of more stochastic parameters.

Artificial intelligence and clinical deterioration

PURPOSE OF REVIEW

To provide an overview of the systems being used to identify and predict clinical deterioration in hospitalised patients, with focus on the current and future role of artificial intelligence (AI).

RECENT FINDINGS

There are five leading AI driven systems in this field: the Advanced Alert Monitor (AAM), the electronic Cardiac Arrest Risk Triage (eCART) score, Hospital wide Alert Via Electronic Noticeboard, the Mayo Clinic Early Warning Score, and the Rothman Index (RI). Each uses Electronic Patient Record (EPR) data and machine learning to predict adverse events. Less mature but relevant evolutions are occurring in the fields of Natural Language Processing, Time and Motion Studies, AI Sepsis and COVID-19 algorithms.

SUMMARY

Research-based AI-driven systems to predict clinical deterioration are increasingly being developed, but few are being implemented into clinical workflows. Escobar et al. (AAM) provide the current gold standard for robust model development and implementation methodology. Multiple technologies show promise, however, the pathway to meaningfully affect patient outcomes remains challenging.

Assessment of the Benefits of Targeted Interventions for Pandemic Control in China Based on Machine Learning Method and Web Service for COVID-19 Policy Simulation

Taking the Chinese city of Xiamen as an example, simulation and quantitative analysis were performed on the transmissions of the Coronavirus Disease 2019 (COVID-19) and the influence of intervention combinations to assist policymakers in the preparation of targeted response measures. A machine learning model was built to estimate the effectiveness of interventions and simulate transmission in different scenarios. The comparison was conducted between simulated and real cases in Xiamen. A web interface with adjustable parameters, including choice of intervention measures, intervention weights, vaccination, and viral variants, was designed for users to run the simulation. The total case number was set as the outcome. The cumulative number was 4,614,641 without restrictions and 78 under the strictest intervention set. Simulation with the parameters closest to the real situation of the Xiamen outbreak was performed to verify the accuracy and reliability of the model. The simulation model generated a duration of 52 days before the daily cases dropped to zero and the final cumulative case number of 200, which were 25 more days and 36 fewer cases than the real situation, respectively. Targeted interventions could benefit the prevention and control of COVID-19 outbreak while safeguarding public health and mitigating impacts on people's livelihood.

Predictive Models for Forecasting Public Health Scenarios: Practical Experiences Applied during the First Wave of the COVID-19 Pandemic

Background: Forecasting the behavior of epidemic outbreaks is vital in public health. This makes it possible to anticipate the planning and organization of the health system, as well as possible restrictive or preventive measures. During the COVID-19 pandemic, this need for prediction has been crucial. This paper attempts to characterize the alternative models that were applied in the first wave of this pandemic context, trying to shed light that could help to understand them for future practical applications. Methods: A systematic literature search was performed in standardized bibliographic repertoires, using keywords and Boolean operators to refine the findings, and selecting articles according to the main PRISMA 2020 statement recommendations. Results: After identifying models used throughout the first wave of this pandemic (between March and June 2020), we begin by examining standard data-driven epidemiological models, including studies applying models such as SIR (Susceptible-Infected-Recovered), SQUIDER, SEIR, time-dependent SIR, and other alternatives. For data-driven methods, we identify experiences using autoregressive integrated moving average (ARIMA), evolutionary genetic programming machine learning, short-term memory (LSTM), and global epidemic and mobility models. Conclusions: The COVID-19 pandemic has led to intensive and evolving use of alternative infectious disease prediction models. At this point it is not easy to decide which prediction method is the best in a generic way. Moreover, although models such as the LSTM emerge as remarkably versatile and useful, the practical applicability of the alternatives depends on the specific context of the underlying variable and on the information of the target to be prioritized. In addition, the robustness of the assessment is conditioned by heterogeneity in the quality of information sources and differences in the characteristics of disease control interventions. Further comprehensive comparison of the performance of models in comparable situations, assessing their predictive validity, is needed. This will help determine the most reliable and practical methods for application in future outbreaks and eventual pandemics.