ABSTRACT
Background. Dengue fever and COVID-19 co-infection constitute a significant public health concern in Latin America, becoming a clinical challenge to distinguish these two entities in early stages of the disease. Clinical outcomes of coinfected hospitalized patients have not been well established. Methods. A cross-sectional study was conducted. We included suspected patients diagnosed with COVID-19/dengue co-infection admitted at Hospital Fundacion Valle del Lili, Cali - Colombia, from March 2020 to March 2021. All dengue patients had positive NS1 and/or IgM dengue antibodies. SARS-CoV-2 infection was confirmed by RT-PCR or antigen rapid test from nasopharyngeal swab. Laboratory and clinical data were recollected from the clinical laboratory database, clinical charts, and institutional COVID-19 registry. Results. A total of 90 COVID-19 patients were included. 72 patients were confirmed only with COVID-19, and 18 with dengue co-infection. Most patients were male: 46 (63.9%) vs. 13 (72.2%). None of these study patients were vaccinated against COVID-19 or dengue. The median time from symptoms onset and the diagnosis was five days, and fever was the most common symptom for both groups. There were significant differences between COVID-19 patients and coinfected patients regarding presence of dyspnea (22.2% vs. 61.1%;p=0.003), desaturation (13.4% vs. 53.3%;p=0.002) and a higher neutrophil/lymphocyte ratio (NLR) (3.84 vs. 5.59;p = 0.038). The co-infection was associated with a worse presentation of the COVID-19 infection (p=0.002), an increased requirement of initial supplemental oxygen therapy (p=0.007), mechanical ventilation (p=0.0004), ICU management at the admission (p=0.002), and ICU final management (p=0.002). Overall mortality in patients with co-infection was 44.4% vs. 6.9% in only COVID-19 infected patients (p< 0.001). Conclusion. Despite the pandemic era, the possibility of co-infection of these two entities must be considered. Admitted coinfected patients were associated with worse clinical outcomes and higher mortality. According to our results, patients with co-infection present with severe respiratory symptoms and an elevated NLR. The impact of the Covid 19 vaccination on this coinfection is unknown.
ABSTRACT
Day-ahead energy forecasting systems struggle to provide accurate demand predictions due to pandemic mitigation measures. Decomposition-Residuals Deep Neural Networks (DR-DNN) are hybrid point-forecasting models that can provide more accurate electricity demand predictions than single models within the COVID-19 era. DR-DNN is a novel two-layer hybrid architecture with: a decomposition and a nonlinear layer. Based on statistical tests, decomposition applies robust signal extraction and filtering of input data into: trend, seasonal and residuals signals. Utilizing calendar information, temporal signals are added: sinusoidal day/night cycles, weekend/weekday, etc. The nonlinear layer learns unknown complex patterns from all those signals, with the usage of well-established deep neural networks. DR-DNN outperformed baselines and state-of-the-art deep neural networks on next-day electricity forecasts within the COVID-19 era (from September 2020 to February 2021), both with fixed and Bayesian optimized hyperparameters. Additionally, model interpretability is improved, by indicating which endogenous or exogenous inputs contribute the most to specific hour-ahead forecasts. Residual signals are very important on the first hour ahead, whereas seasonal patterns on the 24th. Some calendar features also ranked high: whether it is day or night, weekend or weekday and the hour of the day. Temperature was the most important exogenous factor. Author