ObjectiveWe aimed to provide an overview of the psychological status and behavioral consequences of the lockdown during the COVID-19 pandemic in Libya.MethodsA cross-sectional study was conducted among the Libyan population through May and June 2020 in more than 20 cities. The survey comprised basic demographic data of the participants and anxiety symptoms measured using the seven-item Generalized Anxiety Disorder scale with ≥15 as the cut-off score for clinically significant anxiety symptoms. Additionally, a survey regarding the lockdown effect was administered, which (...) consisted of several parts, to measure the lockdown status.ResultsA total of 8084 responses were recorded, of which, 5090 were women and 2994 were men. The mean age for study participants was 27.2 years. Among the participants, 1145 reached the cut-off score to detect anxiety symptoms; however, of the study variables, only five were predictors of clinically significant anxiety: age, gender, marital status, work status, being a financial supporter for the family, and being infected with COVID-19. Women had 1.19 times higher odds to exhibit anxiety symptoms than men. Increasing age was significantly associated with reduced likelihood of exhibiting anxiety symptoms, whereas being married was significantly associated with higher likelihood of anxiety symptoms, compared to not being married. Being suspended from work was associated with an increase in the likelihood of anxiety symptoms. However, we found that being infected with COVID-19 was associated with a 9.59 times higher risk of exhibiting severe anxiety symptoms. Among the study participants, 1451 reported a physical and/or verbal abuse episode from family members, 958 reported abuse outside the family, and 641 reported abuse from enforcers, during the lockdown.ConclusionOur study provided an overview of the psychological and behavioral status, among those who resided in Libya during the civil war and COVID-19 pandemic. The study demonstrates a concerningly high level of clinically significant anxiety during lockdown among the Libyan population during Libya’s lockdown period. (shrink)

The coronavirus disease 2019 pandemic has affected most countries of the world. The detection of Covid-19 positive cases is an important step to fight the pandemic and save human lives. The polymerase chain reaction test is the most used method to detect Covid-19 positive cases. Various molecular methods and serological methods have also been explored to detect Covid-19 positive cases. Machine learning algorithms have been applied to various kinds of datasets to predict Covid-19 positive cases. The machine learning algorithms were (...) applied on a Covid-19 dataset based on commonly taken laboratory tests to predict Covid-19 positive cases. These types of datasets are easy to collect. The paper investigates the application of decision tree ensembles which are accurate and robust to the selection of parameters. As there is an imbalance between the number of positive cases and the number of negative cases, decision tree ensembles developed for imbalanced datasets are applied. F-measure, precision, recall, area under the precision-recall curve, and area under the receiver operating characteristic curve are used to compare different decision tree ensembles. Different performance measures suggest that decision tree ensembles developed for imbalanced datasets perform better. Results also suggest that including age as a variable can improve the performance of various ensembles of decision trees. (shrink)