RESUMO
The sudden outbreak of coronavirus disease 2019 (COVID-19) revealed the need for fast and reliable automatic tools to help health teams. This paper aims to present understandable solutions based on Machine Learning (ML) techniques to deal with COVID-19 screening in routine blood tests. We tested different ML classifiers in a public dataset from the Hospital Albert Einstein, São Paulo, Brazil. After cleaning and pre-processing the data has 608 patients, of which 84 are positive for COVID-19 confirmed by RT-PCR. To understand the model decisions, we introduce (i) a local Decision Tree Explainer (DTX) for local explanation and (ii) a Criteria Graph to aggregate these explanations and portrait a global picture of the results. Random Forest (RF) classifier achieved the best results (accuracy 0.88, F1-score 0.76, sensitivity 0.66, specificity 0.91, and AUROC 0.86). By using DTX and Criteria Graph for cases confirmed by the RF, it was possible to find some patterns among the individuals able to aid the clinicians to understand the interconnection among the blood parameters either globally or on a case-by-case basis. The results are in accordance with the literature and the proposed methodology may be embedded in an electronic health record system.
Assuntos
COVID-19 , Brasil , Árvores de Decisões , Testes Hematológicos , Humanos , Aprendizado de Máquina , SARS-CoV-2RESUMO
OBJETIVO: Investigar alterações dos parâmetros metabólicos no sangue e rim de ratos submetidos à isquemia/reperfusão do membro pélvico. MÉTODOS: Quarenta e oito ratos machos foram distribuídos aleatoriamente em 2 grupos pré-tratados com administração intragástrica de solução salina 2,0 mL (G-1) ou L-alanil-glutamina 0,75 mgKg-1(G-2), uma vez ao dia (7:00h) durante 7 dias. Uma hora após a última gavagem todos os ratos foram anestesiados com éter dietílico, laparotomizados e submetidos ao pinçamento da artéria de ilíaca esquerda, durante 3 horas. Amostras foram coletadas ao término de isquemia e durante a reperfusão (1-3-6h) para determinação das concentrações in vivo de piruvato, lactato, glicose e corpos cetônicos (rim e sangue) e ATP (rim). RESULTADOS: Lactacemia e cetonemia aumentaram no grupo G-2 quando comparadas às aferidas em ratos não-tratados, durante a reperfusão. As concentrações de piruvato diminuíram e de lactato aumentaram significativamente no rim, durante a reperfusão (1h, 3h) em ratos do G-2 comparados aos respectivos controles. Houve um aumento significante nas concentrações renais de glicose, ATP e corpos cetônicos nos ratos tratados com L-alanil-glutamina durante a reperfusão (3h). CONCLUSÕES: A isquemia do membro pélvico em ratos pré-tratados com L-alanil-glutamina induz aumento da lactacemia e da concentração de lactato renal, indicando atividade glicolítica aumentada na medula renal. A hipercetonemia induzida pela oferta do dipeptídeo sugere cetogênese elevada, sinalizada por possível queda nas concentrações plasmáticas de insulina resultante da maior oxidação de glicose e utilização desse hormônio em tecidos periféricos.
Assuntos
Animais , Masculino , Ratos , Dipeptídeos/farmacologia , Membro Posterior/irrigação sanguínea , Rim/efeitos dos fármacos , Traumatismo por Reperfusão/metabolismo , Ácido Láctico/sangue , Ácido Pirúvico/sangue , Cetonas/sangue , Modelos Animais de Doenças , Glicemia/análise , Isquemia/etiologia , Isquemia/metabolismo , Membro Posterior/metabolismo , Distribuição Aleatória , Ratos Wistar , Rim/metabolismo , Estatísticas não Paramétricas , Traumatismo por Reperfusão/sangueRESUMO
PURPOSE: To investigate the effects of l-alanyl-glutamine (Ala-Gln) intragastric administration upon blood and kidney metabolic parameters alterations in rats subjected to ischemia/reperfusion of hind limb. METHODS: Forty-eight male rats were randomized in 2 groups offered via gavage either saline 2.0 mL (G-1) or Ala-Gln solution 0.75 mgKg-1(G-2) once a day at 7 AM during 7 days. One-hour after the last gavage (Day 7) all rats were submitted to ether anesthesia, laparotomy and clamping of the left iliac artery for 3 h. Kidney and blood samples were collected at the end of ischemic period (3 h) and at 1-3-6 h during reperfusion period for metabolites (pyruvate, lactate, glucose and ketone bodies) enzymatic analysis. ATP was also assayed in kidney samples. RESULTS: Lactacemia and ketonemia were significantly increased in Ala-Gln treated rats during reperfusion. Kidney pyruvate concentrations were significantly decreased and tissue lactate concentrations were significantly increased during reperfusion (1 h and 3 h) in G-2 rats compared with respective controls. Glucose, ATP and ketone bodies concentrations were significantly increased in the kidney in L-Ala-Gln treated rats at 3 hours after reperfusion as compared to respective controls. CONCLUSIONS: Unilateral hind limb ischemia in L-Ala-Gln pre-treated rats may induce increased lactacemia and increased kidney lactate concentrations, indicating increased glycolytic activity in renal medulla and in other peripheral tissues. Higher ketonemia during reperfusion may reflect a possible increase in ketogenesis due to lower insulin plasma concentration hepatic signaling as a result of increased glucose oxidation in peripheral tissues, caused by the intra-gastric administration of glutamine dipeptide, suggesting also decreased insulin resistance.
