Effectiveness and trend forecasting of tuberculosis diagnosis after the introduction of GeneXpert in a city in south-eastern Brazil

Machine learning in the loop for tuberculosis diagnosis support

The use of machine learning (ML) for diagnosis support has advanced in the field of health. In the present paper, the results of studying ML techniques in a tuberculosis diagnosis loop in a scenario of limited resources are presented. Data are analyzed using a tuberculosis (TB) therapy program at a health institution in a main city of a developing country using five ML models. Logistic regression, classification trees, random forest, support vector machines, and artificial neural networks are trained under physician supervision following physicians' typical daily work. The models are trained on seven main variables collected when patients arrive at the facility. Additionally, the variables applied to train the models are analyzed, and the models' advantages and limitations are discussed in the context of the automated ML techniques. The results show that artificial neural networks obtain the best results in terms of accuracy, sensitivity, and area under the receiver operating curve. These results represent an improvement over smear microscopy, which is commonly used techniques to detect TB for special cases. Findings demonstrate that ML in the TB diagnosis loop can be reinforced with available data to serve as an alternative diagnosis tool based on data processing in places where the health infrastructure is limited.

Time series forecasting for tuberculosis incidence employing neural network models

Every effort aimed at stopping the expansion of Tuberculosis is important to national programs' struggle to combat this disease. Different computational tools have been proposed in order to design new strategies that allow managing potential patients and thus providing the correct treatment. In this work, artificial neural networks were used for time series forecasting, which were trained with information on reported cases obtained from the national vigilance institution in Colombia. Three neural models were proposed in order to determine the best one according to their forecasting performance. The first approach employed a nonlinear autoregressive model, the second proposal used a recurrent neural network, and the third proposal was based on radial basis functions. The results are presented in terms of the mean average percentage error, which indicates that the models based on traditional methods show better performance compared to connectionist ones. These models contribute to obtaining dynamic information about incidence, thus providing extra-help for health authorities to propose more strategies to control the disease's spread.

Risk-prone territories for spreading tuberculosis, temporal trends and their determinants in a high burden city from São Paulo State, Brazil

Objectives

To identify risk-prone areas for the spread of tuberculosis, analyze spatial variation and temporal trends of the disease in these areas and identify their determinants in a high burden city.

Methods

An ecological study was carried out in Ribeirão Preto, São Paulo, Brazil. The population was composed of pulmonary tuberculosis cases reported in the Tuberculosis Patient Control System between 2006 and 2017. Seasonal Trend Decomposition using the Loess decomposition method was used. Spatial and spatiotemporal scanning statistics were applied to identify risk areas. Spatial Variation in Temporal Trends (SVTT) was used to detect risk-prone territories with changes in the temporal trend. Finally, Pearson's Chi-square test was performed to identify factors associated with the epidemiological situation in the municipality.

Results

Between 2006 and 2017, 1760 cases of pulmonary tuberculosis were reported in the municipality. With spatial scanning, four groups of clusters were identified with relative risks (RR) from 0.19 to 0.52, 1.73, 2.07, and 2.68 to 2.72. With the space–time scan, four clusters were also identified with RR of 0.13 (2008–2013), 1.94 (2010–2015), 2.34 (2006 to 2011), and 2.84 (2014–2017). With the SVTT, a cluster was identified with RR 0.11, an internal time trend of growth (+ 0.09%/year), and an external time trend of decrease (− 0.06%/year). Finally, three risk factors and three protective factors that are associated with the epidemiological situation in the municipality were identified, being: race/brown color (OR: 1.26), without education (OR: 1.71), retired (OR: 1.35), 15 years or more of study (OR: 0.73), not having HIV (OR: 0.55) and not having diabetes (OR: 0.35).

Conclusion

The importance of using spatial analysis tools in identifying areas that should be prioritized for TB control is highlighted, and greater attention is necessary for individuals who fit the profile indicated as “at risk” for the disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12879-022-07500-5.

A Sensitive and Rapid Wastewater Test for SARS-COV-2 and Its Use for the Early Detection of a Cluster of Cases in a Remote Community

Throughout the coronavirus disease 2019 (COVID-19) pandemic, wastewater surveillance has been used to monitor trends in severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) prevalence in the community. A major challenge in establishing wastewater surveillance programs, especially in remote areas, is the need for a well-equipped laboratory for sample analysis. Currently, no options exist for rapid, sensitive, mobile, and easy-to-use wastewater tests for SARS-CoV-2. The performance of the GeneXpert system, which offers cartridge-based, rapid molecular clinical testing for SARS-CoV-2 in a portable platform, was evaluated using wastewater as the input. The GeneXpert demonstrated a SARS-CoV-2 limit of detection in wastewater below 32 copies/mL with a sample processing time of less than an hour. Using wastewater samples collected from multiple sites across Canada during February and March 2021, a high overall agreement (97.8%) was observed between the GeneXpert assay and laboratory-developed tests regarding the presence or absence of SARS-CoV-2. Additionally, with the use of centrifugal filters, the detection threshold of the GeneXpert system was improved to <10 copies/mL in wastewater. Finally, to support on-site wastewater surveillance, GeneXpert testing was implemented in Yellowknife, a remote community in Northern Canada, where its use successfully alerted public health authorities to undetected transmission of COVID-19. The identification of SARS-CoV-2 in wastewater triggered clinical testing of recent travelers and identification of new COVID-19 cases/clusters. Taken together, these results suggest that GeneXpert is a viable option for surveillance of SARS-CoV-2 in wastewater in locations that do not have access to established testing laboratories.

IMPORTANCE Wastewater-based surveillance is a powerful tool that provides an unbiased measure of COVID-19 prevalence in a community. This work describes a sensitive wastewater rapid test for SARS-CoV-2 based on a widely distributed technology, the GeneXpert. The advantages of an easy-to-use wastewater test for SARS-CoV-2 are clear: it supports surveillance in remote communities, improves access to testing, and provides faster results allowing for an immediate public health response. The application of wastewater rapid testing in a remote community facilitated the detection of a COVID-19 cluster and triggered public health action, clearly demonstrating the utility of this technology. Wastewater surveillance will become increasingly important in the postvaccination pandemic landscape as individuals with asymptomatic/mild infections continue transmitting SARS-CoV-2 but are unlikely to be tested.