ABSTRACT
Modeling the dynamics of epidemics helps to propose control strategies based on pharmaceutical and non-pharmaceutical interventions (contact limitation, lockdown, vaccination, etc). Hand-designing such strategies is not trivial because of the number of possible interventions and the difficulty to predict long-term effects. This task can be cast as an optimization problem where state-of-the-art machine learning methods such as deep reinforcement learning might bring significant value. However, the specificity of each domain—epidemic modeling or solving optimization problems—requires strong collaborations between researchers from different fields of expertise. This is why we introduce EpidemiOptim, a Python toolbox that facilitates collaborations between researchers in epidemiology and optimization. EpidemiOptim turns epidemiological models and cost functions into optimization problems via a standard interface commonly used by optimization practitioners (OpenAI Gym). Reinforcement learning algorithms based on Q-Learning with deep neural networks (dqn) and evolutionary algorithms (nsga-ii) are already implemented. We illustrate the use of EpidemiOptim to find optimal policies for dynamical on-off lockdown control under the optimization of the death toll and economic recess using a Susceptible-Exposed-Infectious-Removed (seir) model for COVID-19. Using EpidemiOptim and its interactive visualization platform in Jupyter notebooks, epidemiologists, optimization practitioners and others (e.g. economists) can easily compare epidemiological models, costs functions and optimization algorithms to address important choices to be made by health decision-makers. Trained models can be explored by experts and non-experts via a web interface. ©2021 AI Access Foundation. All rights reserved.
ABSTRACT
BACKGROUND AND OBJECTIVE: Since the first stages of the novel coronavirus 2019 (SARS-CoV-2) outbreak smell and/or taste dysfunction (STD), has been described from 5% to 88% in COVID-19 patients. Objective: we aimed to assess STD in healthcare professionals (HCP), mainly allergists, affected with COVID-19, by means of a survey, and to evaluate the association of STD and their severity with demographic characteristics, symptoms, comorbidities, and hospital admission. METHODS: A 15-item questionnaire was designed including different sections as follows: demographics, diagnostic characteristics, STD patterns, medication use as well as comorbidities. The questionnaire was developed using Google forms, implemented and distributed to members of the Spanish Society of Allergology and Clinical Immunology (SEAIC) and spread via Social Media to be completed by HCP affected with COVID-19. RESULTS: HCP (n=234), 76.5% ≤55 yrs, 73.5% female, completed the survey. There was STD in up to 74.4% of the respondents, 95.6% reporting a moderate-severe impairment. Mean recovery time of taste dysfunction was 21.6±24.0 days in HCP ≤55 yrs and 33.61±26.2 days in >55 yrs (p=0.019). Stratified analysis by severity of STD showed that more than a half of COVID-19 subjects presented severe loss of smell. An older age (>55 yrs) was associated with fever, anorexia, less headache and with a longer persistence of taste dysfunction. CONCLUSIONS: STD is a common symptom in COVID-19, even as a unique or preceding symptom. HCP who declared smell dysfunction (SD) were younger than those not affected with STD. Taste dysfunction (TD) may imply more systemic involvement in COVID-19-positive HCP.