Prediction of SARS-CoV-2-positivity from million-scale complete blood counts using machine learning.

Background: The Complete Blood Count (CBC) is a commonly used low-cost test that measures white blood cells, red blood cells, and platelets in a person's blood. It is a useful tool to support medical decisions, as intrinsic variations of each analyte bring relevant insights regarding potential diseases. In this study, we aimed at developing machine learning models for COVID-19 diagnosis through CBCs, unlocking the predictive power of non-linear relationships between multiple blood analytes. Methods: We collected 809,254 CBCs and 1,088,385 RT-PCR tests for SARS-Cov-2, of which 21% (234,466) were positive, from 900,220 unique individuals. To properly screen COVID-19, we also collected 120,807 CBCs of 16,940 individuals who tested positive for other respiratory viruses. We proposed an ensemble procedure that combines machine learning models for different respiratory infections and analyzed the results in both the first and second waves of COVID-19 cases in Brazil. Results: We obtain a high-performance AUROC of 90 + % for validations in both scenarios. We show that models built solely of SARS-Cov-2 data are biased, performing poorly in the presence of infections due to other RNA respiratory viruses. Conclusions: We demonstrate the potential of a novel machine learning approach for COVID-19 diagnosis based on a CBC and show that aggregating information about other respiratory diseases was essential to guarantee robustness in the results. Given its versatile nature, low cost, and speed, we believe that our tool can be particularly useful in a variety of scenarios-both during the pandemic and after.

No association between vitamin D status and COVID-19 infection in São Paulo, Brazil

ABSTRACT In recent years the immunomodulatory actions of vitamin D, a steroid hormone, have been extensively studied. In 2020, due to the COVID-19 pandemic, the question arose as to 25(OH)D status would be related to susceptibility to SARS-CoV-2 infection, since several studies pointed out a higher prevalence and severity of the disease in populations with low levels of 25(OH)D. Thus, we investigated the 25(OH)D levels in adults "Detected" positive for SARS CoV-2 by RT-PCR (reverse transcriptase polymerase chain reaction) test, and in negative controls, "not Detected", using the Fleury Group's examination database, in Sao Paulo, Brazil. Of a total of 14.692 people with recent assessments of 25(OH)D and RT-PCR tests for COVID-19, 2.345 were positive and 11.585 were negative for the infection. The groups did not differ in the percentage of men and women, or in the age distribution. There were no differences in the distribution of 25(OH)D between the two groups (p = 0.08); mean 25(OH)D of 28.8 ± 21.4 ng/mL and 29.6 ± 18.1 ng/mL, respectively. In the specific population studied, clinical, environmental, socioeconomic and cultural factors should have greater relevance than 25(OH)D in determining the susceptibility to COVID-19.

No association between vitamin D status and COVID-19 infection in São Paulo, Brazil.

In recent years the immunomodulatory actions of vitamin D, a steroid hormone, have been extensively studied. In 2020, due to the COVID-19 pandemic, the question arose as to 25(OH)D status would be related to susceptibility to SARS-CoV-2 infection, since several studies pointed out a higher prevalence and severity of the disease in populations with low levels of 25(OH)D. Thus, we investigated the 25(OH)D levels in adults "Detected" positive for SARS CoV-2 by RT-PCR (reverse transcriptase polymerase chain reaction) test, and in negative controls, "not Detected", using the Fleury Group's examination database, in Sao Paulo, Brazil. Of a total of 14.692 people with recent assessments of 25(OH)D and RT-PCR tests for COVID-19, 2.345 were positive and 11.585 were negative for the infection. The groups did not differ in the percentage of men and women, or in the age distribution. There were no differences in the distribution of 25(OH)D between the two groups (p = 0.08); mean 25(OH)D of 28.8 ± 21.4 ng/mL and 29.6 ± 18.1 ng/mL, respectively. In the specific population studied, clinical, environmental, socioeconomic and cultural factors should have greater relevance than 25(OH)D in determining the susceptibility to COVID-19.