Background COVID-19 is a multisystemic disorder that frequently causes acute kidney injury (AKI). However, the precise clinical and biochemical variables associated with AKI progression in patients with severe COVID-19 remain unclear. Methods We performed a retrospective study on 278 hospitalized patients who were admitted to the ward and intensive care unit (ICU) with COVID-19 between March 2020 and June 2020, at the University Hospital, São Paulo, Brazil. Patients aged ≥ 18 years with COVID-19 confirmed on RT-PCR were included. AKI was defined according to the Kidney Disease Improving Global Outcomes (KDIGO) criteria. We evaluated the incidence of AKI, several clinical variables, medicines used, and outcomes in two sub-groups: COVID-19 patients with AKI (Cov-AKI), and COVID-19 patients without AKI (non-AKI). Univariate and multivariate analyses were performed. Results First, an elevated incidence of AKI (71.2%) was identified, distributed across different stages of the KDIGO criteria. We further observed higher levels of creatinine, C-reactive protein (CRP), leukocytes, neutrophils, monocytes, and neutrophil-to-lymphocyte ratio (NLR) in the Cov-AKI group than in the non-AKI group, at hospital admission. On univariate analysis, Cov-AKI was associated with older age (>62 years), hypertension, CRP, MCV, leucocytes, neutrophils, NLR, combined hydroxychloroquine and azithromycin treatment, use of mechanical ventilation, and vasoactive drugs. Multivariate analysis showed that hypertension and the use of vasoactive drugs were independently associated with a risk of higher AKI in COVID-19 patients. Finally, we preferentially found an altered erythrocyte and leukocyte cellular profile in the Cov-AKI group compared to the non-AKI group, at hospital discharge. Conclusions In our study, the development of AKI in patients with severe COVID-19 was related to inflammatory blood markers and therapy with hydroxychloroquine/azithromycin, with vasopressor requirement and hypertension considered potential risk factors. Thus, attention to the protocol, hypertension, and some blood markers may help assist doctors with decision-making for the management of COVID-19 patients with AKI.

The pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) which is related to new coronavirus disease (COVID-19) has mobilized several scientifics to explore clinical data using soft-computing approaches. In the context of machine learning, previous studies have explored supervised algorithms to predict and support diagnosis based on several clinical parameters from patients diagnosed with and without COVID-19. However, in most of them the decision is based on a "black-box" method, making it impossible to discover the variable relevance in decision making. Hence, in this study, we introduce a non-supervised clustering analysis with neural network self-organizing maps (SOM) as a strategy of decision-making. We propose to identify potential variables in routine blood tests that can support clinician decision-making during COVID-19 diagnosis at hospital admission, facilitating rapid medical intervention. Based on SOM features (visual relationships between clusters and identification of patterns and behaviors), and using linear discriminant analysis , it was possible to detect a group of units of the map with a discrimination power around 83% to SARS-CoV-2-positive patients. In addition, we identified some variables in admission blood tests (Leukocytes, Basophils, Eosinophils, and Red cell Distribution Width) that, in combination had strong influence in the clustering performance, which could assist a possible clinical decision. Thus, although with limitations, we believe that SOM can be used as a soft-computing approach to support clinician decision-making in the context of COVID-19. Keywords Covid-19 diagnostic • SARS-CoV-2 • Self-organizing maps Communicated by Victor Hugo C. de Albuquerque.

Abstract. The automated analysis of programming code developed during IntroduçãoDa mesma forma que em outros contextos do processo de ensino-aprendizagem, avaliar adequadamente o desenvolvimento de competências e habilidades relacionadas ao

Computational thinking has become a required capability in the student learning process, and digital games as a teaching approach have presented promising educational results in the development of this competence. However, properly evaluating the effectiveness and, consequently, student progress in a course using games is still a challenge. One of the most widely implemented ways of evaluation is with an automated analysis of the code developed in the classes during the construction of digital games. Nevertheless, this topic has not yet been explored in aspects such as incremental learning, the model and teaching environment and the influences of acquiring skills and competencies of computational thinking. Motivated by this knowledge gap, this paper introduces a framework proposal to analyze the evolution of computational thinking skills in digital games classes. The framework is based on a data mining technique that aims to facilitate the discovery process of the patterns and behaviors that lead to the acquisition of computational thinking skills, by analyzing clusters with an unsupervised neural network of self-organizing maps (SOM) for this purpose. The framework is composed of a collection of processes and practices structured in data collection, data preprocessing, data analysis, and data visualization. A case study, using Scratch, was executed to validate this approach. The results point to the viability of the framework, highlighting the use of the visual exploratory data analysis, through the SOM maps, as an efficient tool to observe the acquisition of computational thinking skills by the student in an incremental course.

In this paper we perform a visual and exploratory analysis on the data generated by the Dr. Scratch tool from codes produced by students in digital game workshops using Scratch environment.Our findings point to a weak linear relationship between Dr. Scratch's rubric and cyclomatic complexity, since codes with low cyclomatic complexity were developed by students who had better scores in the CT-Test. This may indicate that the use of more advanced Scratch features could be related to the advance of the student's acquisition of Computational Thinking skills. Resumo.Neste artigo realizamos uma análise visual e exploratória dos dados gerados pela ferramenta Dr. Scratch a partir de códigos produzidos por alunos em oficinas de produção de jogos digitais utilizando o ambiente Scratch. Nossas descobertas apontam uma fraca relação linear entre a rubrica do Dr. Scratch e a complexidade ciclomática, visto que códigos com complexidade ciclomática baixa foram desenvolvidos por alunos que obtiveram melhor pontuação no CT-Test, o que pode indicar que o uso dos recursos mais avançados do ambiente Scratch pode estar relacionado com o progresso do aluno na aquisição de competências do Pensamento Computacional.