Validation of Chest Computed Tomography Artificial Intelligence to Determine the Requirement for Mechanical Ventilation and Risk of Mortality in Hospitalized Coronavirus Disease-19 Patients in a Tertiary Care Center In Mexico City

ABSTRACT Background: Artificial intelligence (AI) in radiology has improved diagnostic performance and shortened reading times of coronavirus disease 2019 (COVID-19) patients’ studies. Objectives: The objectives pf the study were to analyze the performance of a chest computed tomography (CT) AI quantitative algorithm for determining the risk of mortality/mechanical ventilation (MV) in hospitalized COVID-19 patients and explore a prognostic multivariate model in a tertiary-care center in Mexico City. Methods: Chest CT images of 166 COVID-19 patients hospitalized from April 1 to 20, 2020, were retrospectively analyzed using AI algorithm software. Data were collected from their medical records. We analyzed the diagnostic yield of the relevant CT variables using the area under the ROC curve (area under the curve [AUC]). Optimal thresholds were obtained using the Youden index. We proposed a predictive logistic model for each outcome based on CT AI measures and predetermined laboratory and clinical characteristics. Results: The highest diagnostic yield of the assessed CT variables for mortality was the percentage of total opacity (threshold >51%; AUC = 0.88, sensitivity = 74%, and specificity = 91%). The AUC of the CT severity score (threshold > 12.5) was 0.88 for MV (sensitivity = 65% and specificity = 92%). The proposed prognostic models include the percentage of opacity and lactate dehydrogenase level for mortality and troponin I and CT severity score for MV requirement. Conclusion: The AI-calculated CT severity score and total opacity percentage showed good diagnostic accuracy for mortality and met MV criteria. The proposed prognostic models using biochemical variables and imaging data measured by AI on chest CT showed good risk classification in our population of hospitalized COVID-19 patients.

Evaluación de las características de la disección aórtica en la población mexicana mediante angiotomografía computarizada / Evaluation of the characteristics of aortic dissection by computed angio-tomography in the Mexican population

Resumen Objetivo: Revisión y análisis de la disección aórtica (DA) en la población mexicana. Método: Revisión retrospectiva de 434 expedientes electrónicos de pacientes con angiotomografía de aorta entre noviembre de 2014 y octubre de 2015. Se obtuvo una muestra de 32 pacientes con diagnóstico de DA de primera vez. Se realizó un análisis de las DA según género, grupo etario, clasificación de Stanford/De Bakey y mortalidad a 6 meses del diagnóstico. Se realizó análisis de significación estadística mediante la Chi-cuadrada para las variables independientes de género, síndrome de Marfan, hipertensión arterial sistémica y enfermedad ateromatosa calcificada en asociación con subtipos, sitios de reentrada y datos de hipoperfusión. Resultados: El 65.6% de los pacientes fueron masculinos, con un promedio de edad de 54.5 años, y el 34.4% fueron femeninos, con un promedio de edad de 42.5 años. El subtipo B/3 fue el más frecuentemente diagnosticado. La tasa de mortalidad a 6 meses fue del 18.7%. Se halló asociación significativa con p marginal en pacientes con síndrome de Marfan y subtipos de DA según Stanford (p = 0.0506), así como asociación significativa en pacientes con aneurisma de aorta abdominal y subtipos de DA según Stanford (p = 0.047104). Conclusiones: La DA es una emergencia en la cual el diagnóstico por imagen y el manejo oportuno son fundamentales para mejorar el pronóstico. En nuestra muestra encontramos asociación significativa de pacientes con antecedente de síndrome de Marfan y aneurisma aórtico abdominal con disecciones según la categoría de Stanford. El resto de las variables independientes no mostraron asociación significativa, en probable relación con el tamaño de la muestra.

Abstract Objective: To review aortic dissection (AD) in the Mexican population. Method: A retrospective study was conducted using 434 medical records of patients with aortic angio-tomography between November 2014 and October 2015. A sample was obtained of 32 patients with a first time diagnosis of AD. An analysis was performed of the dissections according to gender, age group, Stanford/De Bakey classification, and mortality rate 6 months after diagnosis. Statistical analysis was performed by obtaining the Chi squared index for the independent variables of gender, Marfan syndrome, systemic arterial hypertension, as well as calcified atheromatous disease in association with dissection subtypes, re-entry sites, and hypo-perfusion signs. Results: The patients included 65.6% males with a mean age of 54.5 years, and 34.4% females with mean age of 42.5 years. The most common dissection subtype was B/3. Mortality rate at 6 months was 18.7%. There was a significant association, with a marginal P in patients with Marfan syndrome and Stanford subtypes of AD (P = .0506). There was a significant association in patients with abdominal aortic aneurysm, when compared with Stanford subtypes of AD (P = .047104). Conclusions: AD is an emergency in which diagnosis and timely management are essential to improve prognosis. In the sample presented here, a significant association was found in patients with a history of Marfan syndrome and abdominal aneurysms with dissections according to the Stanford classification. The rest of the independent variables did not show any significant association, probably related to the size of the sample.