Big data- and artificial intelligence-based hot-spot analysis of COVID-19: Gauteng, South Africa, as a case study.

The coronavirus disease 2019 (COVID-19) has developed into a pandemic. Data-driven techniques can be used to inform and guide public health decision- and policy-makers. In generalizing the spread of a virus over a large area, such as a province, it must be assumed that the transmission occurs as a stochastic process. It is therefore very difficult for policy and decision makers to understand and visualize the location specific dynamics of the virus on a more granular level. A primary concern is exposing local virus hot-spots, in order to inform and implement non-pharmaceutical interventions. A hot-spot is defined as an area experiencing exponential growth relative to the generalised growth of the pandemic. This paper uses the first and second waves of the COVID-19 epidemic in Gauteng Province, South Africa, as a case study. The study aims provide a data-driven methodology and comprehensive case study to expose location specific virus dynamics within a given area. The methodology uses an unsupervised Gaussian Mixture model to cluster cases at a desired granularity. This is combined with an epidemiological analysis to quantify each cluster's severity, progression and whether it can be defined as a hot-spot.

Addressing the migrant gap: maternal healthcare perspectives on utilising prevention of mother to child transmission (PMTCT) services during the COVID-19 pandemic, South Africa.

BACKGROUND: The COVID-19 pandemic has interrupted the prevention of mother-to-child transmission of HIV (PMTCT) programming in South Africa. In 2020, it was estimated that there were 4 million cross-border migrants in South Africa, some of whom are women living with HIV (WLWH), who are highly mobile and located within peripheral and urban areas of Johannesburg. Little is known about the mobility typologies of these women associated with different movement patterns, the impact of the COVID-19 pandemic on mobility typologies of women utilising PMTCT services and on how changes to services might have affected adherence. OBJECTIVE: To qualitatively explore experiences of different mobility typologies of migrant women utilising PMTCT services in a high mobility context of Johannesburg and how belonging to a specific typology might have affected the health care received and their overall experiences during the COVID-19 pandemic. METHODS: Qualitative semi-structured interviews with 40 pregnant migrant WLWH were conducted from June 2020-June 2021. Participants were recruited through purposive sampling at a public hospital in Johannesburg. A thematic approach was used to analyse interviews. RESULTS: Forty interviews were conducted with 22 cross-border and 18 internal migrants. Women in cross-border migration patterns compared to interprovincial and intraregional mobility experienced barriers of documentation, language availability, mistreatment, education and counselling. Due to border closures, they were unable to receive ART interrupting adherence and relied on SMS reminders to adhere to ART during the pandemic. All 40 women struggled to understand the importance of adherence because of the lack of infrastructure to support social distancing protocols and to provide PMTCT education. CONCLUSIONS: COVID-19 amplified existing challenges for cross-border migrant women to utilise PMTCT services. Future pandemic preparedness should be addressed with differentiated service delivery including multi-month dispensing of ARVs, virtual educational care, and language-sensitive information, responsive to the needs of mobile women to alleviate the burden on the healthcare system.

Estimated SARS-CoV-2 infection rate and fatality risk in Gauteng Province, South Africa: a population-based seroepidemiological survey.

BACKGROUND: Limitations in laboratory testing capacity undermine the ability to quantify the overall burden of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection. METHODS: We undertook a population-based serosurvey for SARS-CoV-2 infection in 26 subdistricts, Gauteng Province (population 15.9 million), South Africa, to estimate SARS-CoV-2 infection, infection fatality rate (IFR) triangulating seroprevalence, recorded COVID-19 deaths and excess-mortality data. We employed three-stage random household sampling with a selection probability proportional to the subdistrict size, stratifying the subdistrict census-sampling frame by housing type and then selecting households from selected clusters. The survey started on 4 November 2020, 8 weeks after the end of the first wave (SARS-CoV-2 nucleic acid amplification test positivity had declined to <10% for the first wave) and coincided with the peak of the second wave. The last sampling was performed on 22 January 2021, which was 9 weeks after the SARS-CoV-2 resurgence. Serum SARS-CoV-2 receptor-binding domain (RBD) immunoglobulin-G (IgG) was measured using a quantitative assay on the Luminex platform. RESULTS: From 6332 individuals in 3453 households, the overall RBD IgG seroprevalence was 19.1% [95% confidence interval (CI): 18.1-20.1%] and similar in children and adults. The seroprevalence varied from 5.5% to 43.2% across subdistricts. Conservatively, there were 2 897 120 (95% CI: 2 743 907-3 056 866) SARS-CoV-2 infections, yielding an infection rate of 19 090 per 100 000 until 9 January 2021, when 330 336 COVID-19 cases were recorded. The estimated IFR using recorded COVID-19 deaths (n = 8198) was 0.28% (95% CI: 0.27-0.30) and 0.67% (95% CI: 0.64-0.71) assuming 90% of modelled natural excess deaths were due to COVID-19 (n = 21 582). Notably, 53.8% (65/122) of individuals with previous self-reported confirmed SARS-CoV-2 infection were RBD IgG seronegative. CONCLUSIONS: The calculated number of SARS-CoV-2 infections was 7.8-fold greater than the recorded COVID-19 cases. The calculated SARS-CoV-2 IFR varied 2.39-fold when calculated using reported COVID-19 deaths (0.28%) compared with excess-mortality-derived COVID-19-attributable deaths (0.67%). Waning RBD IgG may have inadvertently underestimated the number of SARS-CoV-2 infections and conversely overestimated the mortality risk. Epidemic preparedness and response planning for future COVID-19 waves will need to consider the true magnitude of infections, paying close attention to excess-mortality trends rather than absolute reported COVID-19 deaths.

Leveraging Artificial Intelligence and Big Data to Optimize COVID-19 Clinical Public Health and Vaccination Roll-Out Strategies in Africa.

COVID-19 is imposing massive health, social and economic costs. While many developed countries have started vaccinating, most African nations are waiting for vaccine stocks to be allocated and are using clinical public health (CPH) strategies to control the pandemic. The emergence of variants of concern (VOC), unequal access to the vaccine supply and locally specific logistical and vaccine delivery parameters, add complexity to national CPH strategies and amplify the urgent need for effective CPH policies. Big data and artificial intelligence machine learning techniques and collaborations can be instrumental in an accurate, timely, locally nuanced analysis of multiple data sources to inform CPH decision-making, vaccination strategies and their staged roll-out. The Africa-Canada Artificial Intelligence and Data Innovation Consortium (ACADIC) has been established to develop and employ machine learning techniques to design CPH strategies in Africa, which requires ongoing collaboration, testing and development to maximize the equity and effectiveness of COVID-19-related CPH interventions.

Development of an Early Alert System for an Additional Wave of COVID-19 Cases Using a Recurrent Neural Network with Long Short-Term Memory.

The impact of the still ongoing "Coronavirus Disease 2019" (COVID-19) pandemic has been and is still vast, affecting not only global human health and stretching healthcare facilities, but also profoundly disrupting societal and economic systems worldwide. The nature of the way the virus spreads causes cases to come in further recurring waves. This is due a complex array of biological, societal and environmental factors, including the novel nature of the emerging pathogen. Other parameters explaining the epidemic trend consisting of recurring waves are logistic-organizational challenges in the implementation of the vaccine roll-out, scarcity of doses and human resources, seasonality, meteorological drivers, and community heterogeneity, as well as cycles of strengthening and easing/lifting of the mitigation interventions. Therefore, it is crucial to be able to have an early alert system to identify when another wave of cases is about to occur. The availability of a variety of newly developed indicators allows for the exploration of multi-feature prediction models for case data. Ten indicators were selected as features for our prediction model. The model chosen is a Recurrent Neural Network with Long Short-Term Memory. This paper documents the development of an early alert/detection system that functions by predicting future daily confirmed cases based on a series of features that include mobility and stringency indices, and epidemiological parameters. The model is trained on the intermittent period in between the first and the second wave, in all of the South African provinces.