Alexandra A. de Souza

[This retracts the article DOI: 10.1007/s00500-021-05810-5.].

Data classification is an automatic or semi-automatic process that, utilizing artificial intelligence algorithms, learns the variable and class relationships of a dataset for use a posteriori in situations where the class result is unknown. For many years, work on this topic has been aimed at increasing the hit rates of algorithms. However, when the problem is restricted to applications in healthcare, besides the concern with performance, it is also necessary to design algorithms whose results are understandable by the specialists responsible for making the decisions. Among the problems in the field of medicine, a current focus is related to COVID-19: AI algorithms may contribute to early diagnosis. Among the available COVID-19 data, the blood test is a typical procedure performed when the patient seeks the hospital, and its use in the diagnosis allows reducing the need for other diagnostic tests that can impact the detection time and add to costs. In this work, we propose using self-organizing map (SOM) to discover attributes in blood test examinations that are relevant for COVID-19 diagnosis. We applied SOM and an entropy calculation in the definition of a hierarchical, semi-supervised and explainable model named TESSOM (tree-based entropy-structured self-organizing maps), in which the main feature is enhancing the investigation of groups of cases with high levels of class overlap, as far as the diagnostic outcome is concerned. Framing the TESSOM algorithm in the context of explainable artificial intelligence (XAI) makes it possible to explain the results to an expert in a simplified way. It is demonstrated in the paper that the use of the TESSOM algorithm to identify attributes of blood tests can help with the identification of COVID-19 cases. It providing a performance increase in 1.489% in multiple scenarios when analyzing 2207 cases from three hospitals in the state of São Paulo, Brazil. This work is a starting point for researchers to identify relevant attributes of blood tests for COVID-19 and to support the diagnosis of other diseases.

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.

A análise automatizada de código desenvolvido por alunos em atividades de desenvolvimento do Pensamento Computacional vem mostrando potencial em identificar práticas e estratégias utilizadas para a resolução dos problemas propostos. No entanto, ainda não tem sido explorada em detalhes a evolução do código produzido por alunos e sua relação com a aquisição de competências do Pensamento Computacional. Neste estudo foram empregados mapas auto-organizáveis na análise de jogos produzidos por alunos em uma Oficina de Produção de Jogos Digitais nos anos de 2013 e 2017. O resultado do processo de treinamento produziu agrupamentos no mapa que apresentaram coerência com características dos jogos vinculadas a estratégias utilizadas pelos alunos que foram previamente observadas em sala de aula. Implicações dos resultados para estudos futuros também são discutidas.