Covid-19 Automated Diagnosis and Risk Assessment through Metabolomics and Machine Learning.

COVID-19 is still placing a heavy health and financial burden worldwide. Impairment in patient screening and risk management plays a fundamental role on how governments and authorities are directing resources, planning reopening, as well as sanitary countermeasures, especially in regions where poverty is a major component in the equation. An efficient diagnostic method must be highly accurate, while having a cost-effective profile. We combined a machine learning-based algorithm with mass spectrometry to create an expeditious platform that discriminate COVID-19 in plasma samples within minutes, while also providing tools for risk assessment, to assist healthcare professionals in patient management and decision-making. A cross-sectional study enrolled 815 patients (442 COVID-19, 350 controls and 23 COVID-19 suspicious) from three Brazilian epicenters from April to July 2020. We were able to elect and identify 19 molecules related to the disease's pathophysiology and several discriminating features to patient's health-related outcomes. The method applied for COVID-19 diagnosis showed specificity >96% and sensitivity >83%, and specificity >80% and sensitivity >85% during risk assessment, both from blinded data. Our method introduced a new approach for COVID-19 screening, providing the indirect detection of infection through metabolites and contextualizing the findings with the disease's pathophysiology. The pairwise analysis of biomarkers brought robustness to the model developed using machine learning algorithms, transforming this screening approach in a tool with great potential for real-world application.

Use of anthropophilic culicid-based xenosurveillance as a proxy for Plasmodium vivax malaria burden and transmission hotspots identification.

Vector-borne diseases account for more than 17% of all infectious diseases, causing more than one million deaths annually. Malaria remains one of the most important public health problems worldwide. These vectors are bloodsucking insects, which can transmit disease-producing microorganisms during a blood meal. The contact of culicids with human populations living in malaria-endemic areas suggests that the identification of Plasmodium genetic material in the blood present in the gut of these mosquitoes may be possible. The process of assessing the blood meal for the presence of pathogens is termed 'xenosurveillance'. In view of this, the present work investigated the relationship between the frequency with which Plasmodium DNA is found in culicids and the frequency with which individuals are found to be carrying malaria parasites. A cross-sectional study was performed in a peri-urban area of Manaus, in the Western Brazilian Amazon, by simultaneously collecting human blood samples and trapping culicids from households. A total of 875 individuals were included in the study and a total of 13,374mosquito specimens were captured. Malaria prevalence in the study area was 7.7%. The frequency of households with at least one culicid specimen carrying Plasmodium DNA was 6.4%. Plasmodium infection incidence was significantly related to whether any Plasmodium positive blood-fed culicid was found in the same household [IRR 3.49 (CI95% 1.38-8.84); p = 0.008] and for indoor-collected culicids [IRR 4.07 (CI95%1.25-13.24); p = 0.020]. Furthermore, the number of infected people in the house at the time of mosquito collection was related to whether there were any positive blood-fed culicid mosquitoes in that household for collection methods combined [IRR 4.48 (CI95%2.22-9.05); p<0.001] or only for indoor-collected culicids [IRR 4.88 (CI95%2.01-11.82); p<0.001]. Our results suggest that xenosurveillance can be used in endemic tropical regions in order to estimate the malaria burden and identify transmission foci in areas where Plasmodium vivax is predominant.