No abstract

The term machine learning refers to a collection of tools used for identifying patterns in data. As opposed to traditional methods of pattern identification, machine learning tools relies on artificial intelligence to map out patters from large amounts of data, can self-improve as and when new data becomes available and is quicker in accomplishing these tasks. This review describes various techniques of machine learning that have been used in the past in the prediction, detection and management of infectious diseases, and how these tools are being brought into the battle against COVID-19. In addition, we also discuss their applications in various stages of the pandemic, the advantages, disadvantages and possible pit falls.

To determine if D-dimers are elevated in individuals with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection who have adverse clinical outcomes including all-cause mortality, intensive care unit (ICU) admission or acute respiratory distress syndrome (ARDS). Methods: We conducted a systematic review and meta-analysis of the published literature in PubMed, Embase and Cochrane databases through April 9, 2020 for studies evaluating D-dimer levels in SARS-COV-2 infected patients with and without a composite clinical endpoint, defined as the presence of all-cause of mortality, Intensive care unit (ICU) admission or acute respiratory distress syndrome (ARDS). A total of six studies were included in the meta-analysis. Results: D-dimers were significantly increased in patients with the composite clinical end point than in those without (SMD, 1.67 ug/ml (95% CI, 0.72-2.62 ug/ml). The SMD of the studies (Tang et al, Zhou et al, Chen et al), which used only mortality as an outcome measure was 2.5 ug/mL (95% CI, 0.62-4.41 ug/ml). Conclusion: We conclude that SARS-CoV-2 infected patients with elevated D-dimers have worse clinical outcomes (all-cause mortality, ICU admission or ARDS) and thus measurement of D-dimers can guide in clinical decision making.

Tocilizumab is an interleukin receptor inhibitor that has been used in patients with COVID-19 pneumonia. There are recent randomized controlled trials (RCTs) that evaluated the efficacy and safety of tocilizumab in hospitalized patients with COVID-19 pneumonia. We performed a systematic review and meta-analysis of RCTs that evaluated the effectiveness of tocilizumab in hospitalized patients with COVID-19 not requiring mechanical ventilation. RCTs comparing tocilizumab with the standard of care treatment in hospitalized patients with COVID-19 pneumonia not requiring mechanical ventilation at the time of administration were included for analysis. The primary outcome was a composite of mechanical ventilation or 28-day mortality and the secondary outcomes were 28-day mortality and major adverse events. A total of 6 RCTs were included for the analysis. Tocilizumab was associated with a statistically significant reduction in the primary composite outcome of mechanical ventilation or 28-day mortality (risk ratio (RR): 0.83 (95% CI: 0.74 to 0.92, I2=0, tau2=0). Treatment with tocilizumab did not show a statistically significant reduction in 28-day mortality (RR: 0.90 (95% CI: 0.76 to 1.07), I2=0, tau2=0) and rate of serious adverse events (RR: 0.82 (95% CI: 0.62 to 1.10), I2=0, tau2=0). Tocilizumab was associated with a decrease in the incidence of primary outcome, that is, mechanical ventilation or death at 28 days in hospitalized patients with COVID-19 pneumonia.

No abstract