Use of machine learning to assess the prognostic utility of radiomic features for in-hospital COVID-19 mortality.

As portable chest X-rays are an efficient means of triaging emergent cases, their use has raised the question as to whether imaging carries additional prognostic utility for survival among patients with COVID-19. This study assessed the importance of known risk factors on in-hospital mortality and investigated the predictive utility of radiomic texture features using various machine learning approaches. We detected incremental improvements in survival prognostication utilizing texture features derived from emergent chest X-rays, particularly among older patients or those with a higher comorbidity burden. Important features included age, oxygen saturation, blood pressure, and certain comorbid conditions, as well as image features related to the intensity and variability of pixel distribution. Thus, widely available chest X-rays, in conjunction with clinical information, may be predictive of survival outcomes of patients with COVID-19, especially older, sicker patients, and can aid in disease management by providing additional information.

Analysis of cardiovascular disease factors on SARS-CoV-2 infection severity.

Background: At present, COVID-19 is a global pandemic and is seriously harmful to humans. In this retrospective study, the aim was to investigate the interaction between CVD and COVID-19. Methods: A total of 180 patients diagnosed with COVID-19 in Yichang Central People's Hospital from 29 January to 17 March 2020 were initially included. The medical history, clinical manifestations at the time of admission, laboratory test results, hospitalization time and complications were recorded. According to the medical history, the patients were assigned to the nonsevere group with non-CVD (n = 90), the nonsevere group with CVD (n = 22), the severe group with non-CVD (n = 40) and the severe group with CVD (n = 28). Results: In the severe group, compared with non-CVD patients, CVD patients had a significantly higher incidence of fever (P < 0.05). However, compared with the nonsevere group, the severe group had significantly higher proportions of patients with hypertension, type 2 diabetes mellitus, CHD and HF (all P < 0.05). Among the patients with nonsevere COVID-19, the WBC count and the levels of IL-6, CRP, D-dimer, NT-proBNP, and FBG were significantly higher and the Hb level was significantly lower in the CVD patients than in the non-CVD patients (all P < 0.05). However, among the patients with severe COVID-19, only the level of NT-proBNP was significantly higher in CVD patients than in non-CVD patients (P < 0.05). In addition, the WBC count and the levels of IL-6, CRP, D-dimer, CKMB, ALT, AST, SCR, NT-proBNP, and FBG were significantly higher and the Hb level was significantly lower in the severe group than in the nonsevere group (all P < 0.05). However, among the patients with severe COVID-19, the incidences of acute myocardial injury, acute kidney injury, arrhythmia, and sudden death were significantly higher in the CVD group than in the non-CVD group (all P < 0.05). The same results were found in the comparison of the nonsevere group with the severe group. Among the patients with nonsevere COVID-19, those without CVD had a mean hospitalization duration of 25.25 (SD 7.61) days, while those with CVD had a mean hospitalization duration of 28.77 (SD 6.11) days; the difference was significant (P < 0.05). The same results were found in the comparison of the severe group. Conclusions: CVD affects the severity of COVID-19. COVID-19 also increases the risk of severe CVD.

Antecedentes: La infección por SARS-CoV-2 está provocando graves consecuencias en la humanidad. El objetivo de este estudio retrospectivo fue investigar el impacto de las enfermedades cardiovasculares (ECV) en la gravedad de dicha infección. Métodos: Entre el 29 de enero y el 17 de marzo de 2020, se diagnosticaron 180 pacientes con neumonía por SARS-CoV-2 en el Hospital Popular Central de Yichang. Se registraron los antecedentes, manifestaciones clínicas, resultados de laboratorio, tiempo de hospitalización y complicaciones. Los pacientes se dividieron en cuatro grupos: 1) infección no grave sin ECV (n = 90), 2) infección no grave con ECV (n = 22), 3) infección grave sin ECV (n = 40) y 4) infección grave con ECV (n = 28). Resultados: La prevalencia de fiebre en los pacientes con ECV fue significativamente mayor que en aquellos sin ECV (P < 0,05). Sin embargo, en comparación con los pacientes no graves, la proporción de pacientes con hipertensión, diabetes mellitus tipo 2, cardiopatía coronaria e insuficiencia cardíaca en los pacientes graves fue significativamente mayor (p< 0,05). Los niveles de recuento de leucocitos, IL-6, PCR, dímero D, NT-proBNP y glucemia en ayunas (GA) en pacientes con ECV fueron significativamente mayores que en los de pacientes sin ECV, aunque los niveles de Hb fueron significativamente menores que los de los pacientes sin ECV (p< 0,05). Sin embargo, los valores de NT-proBNP en pacientes con ECV fueron significativamente mayores que en los pacientes sin ECV (P< 0,05). Además, el recuento de leucocitos y los niveles de IL-6, PCR, dímero D, CK-MB, ALT, AST, creatinina, NT-proBNPy GA en el grupo de pacientes graves fueron significativamente mayores que en el grupo no grave, mientras que los valores de Hb fueron significativamente menores que en el grupo no grave (p< 0,05). La prevalencia de lesión miocárdica aguda, lesión renal aguda, arritmia y muerte súbita en el grupo con ECV fue significativamente mayor que en el grupo sin ECV (p< 0,05). Los mismos resultados se encontraron al comparar los pacientes no graves con aquellos con infección grave. Entre los pacientes no graves, la duración media de la estancia hospitalaria fue de 25,25 (DE: 7,61) días en los pacientes sin ECV, mientras que la duración media de la estancia hospitalaria fue de 28,77 (DE: 6,11) días en los pacientes con ECV (p< 0,05). Los mismos resultados se observaron al comparar los dos grupos con infección grave. Conclusiones: La infección por SARS-CoV-2 es de evolución más grave en los pacientes con ECV.

Analysis of cardiovascular disease factors on SARS-CoV-2 infection severity.

BACKGROUND: At present, COVID-19 is a global pandemic and is seriously harmful to humans. In this retrospective study, the aim was to investigate the interaction between CVD and COVID-19. METHODS: A total of 180 patients diagnosed with COVID-19 in Yichang Central People's Hospital from 29 January to 17 March 2020 were initially included. The medical history, clinical manifestations at the time of admission, laboratory test results, hospitalization time and complications were recorded. According to the medical history, the patients were assigned to the nonsevere group with non-CVD (n=90), the nonsevere group with CVD (n=22), the severe group with non-CVD (n=40) and the severe group with CVD (n=28). RESULTS: In the severe group, compared with non-CVD patients, CVD patients had a significantly higher incidence of fever (P<0.05). However, compared with the nonsevere group, the severe group had significantly higher proportions of patients with hypertension, type 2 diabetes mellitus, CHD and HF (all P<0.05). Among the patients with nonsevere COVID-19, the WBC count and the levels of IL-6, CRP, D-dimer, NT-proBNP, and FBG were significantly higher and the Hb level was significantly lower in the CVD patients than in the non-CVD patients (all P<0.05). However, among the patients with severe COVID-19, only the level of NT-proBNP was significantly higher in CVD patients than in non-CVD patients (P<0.05). In addition, the WBC count and the levels of IL-6, CRP, D-dimer, CKMB, ALT, AST, SCR, NT-proBNP, and FBG were significantly higher and the Hb level was significantly lower in the severe group than in the nonsevere group (all P<0.05). However, among the patients with severe COVID-19, the incidences of acute myocardial injury, acute kidney injury, arrhythmia, and sudden death were significantly higher in the CVD group than in the non-CVD group (all P<0.05). The same results were found in the comparison of the nonsevere group with the severe group. Among the patients with nonsevere COVID-19, those without CVD had a mean hospitalization duration of 25.25 (SD 7.61) days, while those with CVD had a mean hospitalization duration of 28.77 (SD 6.11) days; the difference was significant (P<0.05). The same results were found in the comparison of the severe group. CONCLUSIONS: CVD affects the severity of COVID-19. COVID-19 also increases the risk of severe CVD.

Clinical characteristics and viral shedding kinetics of 38 asymptomatic patients with coronavirus disease 2019: A retrospective observational study.

ABSTRACT: This study aims to investigate the clinical characteristics and viral shedding kinetics of asymptomatic patients with coronavirus disease 2019 (COVID-19).The data of 38 asymptomatic patients positive for SARS-CoV-2 nucleic acid were collected from February to March 2020 in Tuanfeng County, Huanggang, Hubei, China. The epidemiology, laboratory examination, chest imaging, viral nucleic acid test results, clinical characteristics, and viral shedding time were summarized in this retrospective study.The study included 20 family members of patients with COVID-19, 10 medical personnel participating in COVID-19 treatment or working in a fever clinic, 6 personnel from quarantine places, 1 individual with a close contact history with confirmed patients, and 1 local epidemic prevention personnel. All were positive for SARS-CoV-2 nucleic acid. The white blood cell (WBC) count, the absolute value of lymphocytes, C-reactive protein (CRP), and D-dimer were normal. Pneumonia manifestations were not found in the chest computed tomography (CT) scan of 36 patients; the remaining 2 cases included a 1-year-old child and a pregnant woman, and they did not undergo chest CT. The viral shedding time was 6 days.All asymptomatic patients with COVID-19 had a history of close contact or exposure. Laboratory tests were normal. Chest imaging did not show any pneumonia manifestation. The viral shedding time was <10 days, which is shorter than that of patients with COVID-19. A timely discovery of such asymptomatic infections is crucial for blocking the spread of the virus and strengthening the prevention and control measures.