Forecasting SARS-CoV-2 transmission and clinical risk at small spatial scales by the application of machine learning architectures to syndromic surveillance data

Ward, Thomas; Johnsen, Alexander; Ng, Stanley; Chollet, François

doi:10.1038/s42256-022-00538-9

Cited by 11 publications

(6 citation statements)

References 47 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Existing work on COVID indicator analysis at fine spatial scales explores cases [ 1 ], which we expand upon to show that there are other novel, viable indicators at these scales. Building on the data collection approach used to determine clinical risk at low spatial geographies [ 6 ], this analysis shows the Google Trends data’s utility can be extended to hospital admissions at a hospital Trust, rather than administrative geography. While also analysing leading indicators of hospital admissions, we extend the work of forecasting US hospital admissions in 2021 [ 40 ] to show effectiveness of Google Trends as a leading indicator within the Omicron waves.…”

Section: Discussionmentioning

confidence: 99%

“…A large set of potentially predictive Google Trends search terms were analysed in previous research to determine which terms should be evaluated [ 6 ]. Initially, over 1,000 terms were captured from the most common phrases used within NHS Pathway 111 telephonic COVID-19 triages, COVID-19 symptoms, over-the-counter medicine, and natural language variations on requests for tests.…”

Section: Datamentioning

confidence: 99%

“…From the wide range of potential terms, 84 COVID-19-relevant Google Trends search term volume scores were collected hourly and aggregated to daily scores per lower tier local authorities (LTLAs) across the UK [ 6 ]. An LTLA is an administrative geography within the UK, capturing a district of local government, smaller than a county.…”

Section: Datamentioning

confidence: 99%

“…There has been work exploring indirect surveillance approaches for disease incidence. For example, perturbations in Google Trend’s search terms have been shown to precede cases and deaths at a national level [ 5 ] and incorporated in neural network architectures to forecast clinical risk in the UK [ 6 ]. These Google Trend search queries also work with other epidemic metrics of interest by improving model predictive performance of case rates, hotspot detection [ 7 ], and deaths at a US state level [ 8 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Understanding the leading indicators of hospital admissions from COVID-19 across successive waves in the UK

Mellor,

Overton,

Fyles

et al. 2023

Epidemiol. Infect.

Self Cite

View full text Add to dashboard Cite

Following the end of universal testing in the UK, hospital admissions are a key measure of COVID-19 pandemic pressure. Understanding leading indicators of admissions at the National Health Service (NHS) Trust, regional and national geographies help health services plan for ongoing pressures. We explored the spatio-temporal relationships of leading indicators of hospitalisations across SARS-CoV-2 waves in England. This analysis includes an evaluation of internet search volumes from Google Trends, NHS triage calls and online queries, the NHS COVID-19 app, lateral flow devices (LFDs), and the ZOE app. Data sources were analysed for their feasibility as leading indicators using Granger causality, cross-correlation, and dynamic time warping at fine spatial scales. Google Trends and NHS triages consistently temporally led admissions in most locations, with lead times ranging from 5 to 20 days, whereas an inconsistent relationship was found for the ZOE app, NHS COVID-19 app, and LFD testing, which diminished with spatial resolution, showing cross-correlation of leads between –7 and 7 days. The results indicate that novel surveillance sources can be used effectively to understand the expected healthcare burden within hospital administrative areas though the temporal and spatial heterogeneity of these relationships is a key determinant of their operational public health utility.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Datamentioning

confidence: 99%

Section: Datamentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Understanding the leading indicators of hospital admissions from COVID-19 across successive waves in the UK

Mellor,

Overton,

Fyles

et al. 2023

Epidemiol. Infect.

Self Cite

View full text Add to dashboard Cite

show abstract

“…RT-PCR testing became available for all symptomatic individuals in the UK from May 2020 [3] and more accessible across the country as laboratory infrastructure was developed [4]. Geographic and socio-economic disparities in test availability were apparent at the start of the pandemic [5,6] and this became less significant with the Home Test Service (HTS) that delivered RT-PCR tests through the postal service. In October 2020, LFTs became available in a limited capacity, and began to be used more widely in April 2021 within institutional settings and for asymptomatic infections [7].…”

Section: Introductionmentioning

confidence: 99%

Bayesian spatial modelling of localised SARS-CoV-2 transmission through mobility networks across England

Ward,

Morris,

Gelman

et al. 2023

PLoS Comput Biol

View full text Add to dashboard Cite

In the early phases of growth, resurgent epidemic waves of SARS-CoV-2 incidence have been characterised by localised outbreaks. Therefore, understanding the geographic dispersion of emerging variants at the start of an outbreak is key for situational public health awareness. Using telecoms data, we derived mobility networks describing the movement patterns between local authorities in England, which we have used to inform the spatial structure of a Bayesian BYM2 model. Surge testing interventions can result in spatio-temporal sampling bias, and we account for this by extending the BYM2 model to include a random effect for each timepoint in a given area. Simulated-scenario modelling and real-world analyses of each variant that became dominant in England were conducted using our BYM2 model at local authority level in England. Simulated datasets were created using a stochastic metapopulation model, with the transmission rates between different areas parameterised using telecoms mobility data. Different scenarios were constructed to reproduce real-world spatial dispersion patterns that could prove challenging to inference, and we used these scenarios to understand the performance characteristics of the BYM2 model. The model performed better than unadjusted test positivity in all the simulation-scenarios, and in particular when sample sizes were small, or data was missing for geographical areas. Through the analyses of emerging variant transmission across England, we found a reduction in the early growth phase geographic clustering of later dominant variants as England became more interconnected from early 2022 and public health interventions were reduced. We have also shown the recent increased geographic spread and dominance of variants with similar mutations in the receptor binding domain, which may be indicative of convergent evolution of SARS-CoV-2 variants.

show abstract

Forecasting healthcare service volumes with machine learning algorithms

Yang,

Zhu,

Wang

2024

Journal of Forecasting

View full text Add to dashboard Cite

As an efficacious solution to remedying the imbalance of medical resources, the online medical platform has burgeoned expeditiously. Apt allotment of medical resources on the medical platform can facilitate patients in reasonably selecting physicians and time slots, coordinating doctors' clinical arrangements, and generating precise projections of medical platform service volume to enhance patient satisfaction and alleviate physicians' workload. To this end, grounded in the data‐driven method, this paper assembles an exhaustive feature set encompassing hospital features, physician features, and patient features. Through feature selection, appropriate features are screened, and machine learning algorithms are leveraged to accurately forecast doctors' online consultation volume. Subsequently, to glean the influence relationship between online medical services and offline medical services, this paper introduces features of offline medical services such as hospital registration volume and regional gross domestic product (GDP) to solve the prediction of offline medical service volume using online medical information. The findings signify that online data feature prediction can pinpoint superior machine learning models for online medical platform service volume (with the optimal accuracy up to 96.89%). Online features exert a positive effect on predicting offline medical service volume, but the accuracy declines to some degree (the optimal accuracy is 73%). Physicians with favorable reputations on the online platform are more susceptible to attain higher offline appointment volumes when online consultation volume is a vital feature impacting offline appointment volume.

show abstract

Forecasting SARS-CoV-2 transmission and clinical risk at small spatial scales by the application of machine learning architectures to syndromic surveillance data

Cited by 11 publications

References 47 publications

Understanding the leading indicators of hospital admissions from COVID-19 across successive waves in the UK

Understanding the leading indicators of hospital admissions from COVID-19 across successive waves in the UK

Bayesian spatial modelling of localised SARS-CoV-2 transmission through mobility networks across England

Forecasting healthcare service volumes with machine learning algorithms

Contact Info

Product

Resources

About