Objetivo. Estudiar la factibilidad de utilización de la inteligencia artificial como método sensible y específico para el cribado de COVID-19 en pacientes con afecciones respiratorias empleando imágenes de tórax obtenidas con tomógrafo y una plataforma de telemedicina. Métodos. Entre marzo del 2020 y junio del 2021 se realizó un estudio observacional descriptivo multicéntrico de factibilidad basada en inteligencia artificial (IA) para el cribado de COVID-19 en imágenes de tórax de pacientes con afecciones respiratorias que acudieron a hospitales públicos. El diagnóstico de las imágenes tomográficas de tórax se realizó a través de la plataforma de IA; luego, se comparó con el diagnóstico molecular (RT-PCR) para determinar la concordancia entre ambos y analizar su factibilidad para el cribado de pacientes con sospecha de COVID-19. Las imágenes y los resultados diagnóstico se enviaron a través de una plataforma de telemedicina. Resultados. Se realizó el cribado de 3 514 pacientes con sospecha diagnóstica de COVID-19, en 14 hospitales a nivel nacional. La mayoría de los pacientes tenían entre 27 y 59 años, seguidos por los mayores de 60 años. La edad promedio fue de 48,6 años; el 52,8% eran de sexo masculino. Los hallazgos más frecuentes fueron neumonía grave, neumonía bilateral con derrame pleural, enfisema pulmonar bilateral y opacidad difusa en vidrio esmerilado, entre otros. Se determinó un promedio de 93% de concordancia y 7% de discordancia entre las imágenes analizadas mediante IA y la RT-PCR. La sensibilidad y especificidad del sistema de IA, obtenidas comparando el resultado del cribado obtenido por IA con la RT-PCR, fueron de 93% y 80% respectivamente. Conclusiones. Es viable la utilización de IA sensible y específica para la detección rápida estratificada de COVID-19 en pacientes con afecciones respiratorias utilizando imágenes obtenidas mediante tomografía de tórax y una plataforma de telemedicina en los hospitales públicos de Paraguay.

Aim: The aim of the study was to present the results and impact of the application of artificial intelligence (AI) in the rapid diagnosis of COVID-19 by telemedicine in public health in Paraguay. Methods: This is a descriptive, multi-centered, observational design feasibility study based on an AI tool for the rapid detection of COVID-19 in chest computed tomography (CT) images of patients with respiratory difficulties attending the country’s public hospitals. The patients’ digital CT images were transmitted to the AI diagnostic platform, and after a few minutes, radiologists and pneumologists specialized in COVID-19 downloaded the images for evaluation, confirmation of diagnosis, and comparison with the genetic diagnosis (reverse transcription polymerase chain reaction (RT-PCR)). It was also determined the percentage of agreement between two similar AI systems applied in parallel to study the viability of using it as an alternative method of screening patients with COVID-19 through telemedicine. Results: Between March and August 2020, 911 rapid diagnostic tests were carried out on patients with respiratory disorders to rule out COVID-19 in 14 hospitals nationwide. The average age of patients was 50.7 years, 62.6% were male and 37.4% female. Most of the diagnosed respiratory conditions corresponded to the age group of 27–59 years (252 studies), the second most frequent corresponded to the group over 60 years, and the third to the group of 19–26 years. The most frequent findings of the radiologists/pneumologists were severe pneumonia, bilateral pneumonia with pleural effusion, bilateral pulmonary emphysema, diffuse ground glass opacity, hemidiaphragmatic paresis, calcified granuloma in the lower right lobe, bilateral pleural effusion, sequelae of tuberculosis, bilateral emphysema, and fibrotic changes, among others. Overall, an average of 86% agreement and 14% diagnostic discordance was determined between the two AI systems. The sensitivity of the AI system was 93% and the specificity 80% compared with RT-PCR. Conclusion: Paraguay has an AI-based telemedicine screening system for the rapid stratified detection of COVID-19 from chest CT images of patients with respiratory conditions. This application strengthens the integrated network of health services, rationalizing the use of specialized human resources, equipment, and inputs for laboratory diagnosis.

Crítica al estudio factibilidad de la utilización de la inteligencia artificial para el cribado de pacientes con COVID-19 en ParaguayEste es un artículo de acceso abierto distribuido bajo los términos de la licencia Creative Commons Attribution-NonCommercial-NoDerivs 3.0 IGO, que permite su uso, distribución y reproducción en cualquier medio, siempre que el trabajo original se cite de la manera adecuada. No se permiten modificaciones a los artículos ni su uso comercial. Al reproducir un artículo no debe haber ningún indicio de que la OPS o el artículo avalan a una organización o un producto específico. El uso del logo de la OPS no está permitido. Esta leyenda debe conservarse, junto con la URL original del artículo. Crédito del logo y texto open access: PLoS, bajo licencia Creative Commons Attribution-Share Alike 3.0 Unported.

IntroductionArtificial intelligence (AI) and innovative technology offer opportunities for enhanced health care during the COVID-19 pandemic. Populations living in low-income countries do not have access to reverse transcription polymerase chain reaction (RT-PCR) testing for COVID-19 and, thus, depend on the scarce resources of their health system. In this context, an automated screening system for COVID-19 based on AI for a telemedicine platform could be directed towards alleviating the current lack of trained radiologists who can interpret computed tomography images at countryside hospitals.MethodsThis descriptive study was carried out in Paraguay by the Telemedicine Unit of the Ministry of Public Health and Social Welfare in collaboration with the Department of Biomedical Engineering and Imaging of the Health Science Research Institute and the University of the Basque Country. The utility of the screening system for COVID-19 was analyzed by dividing the results from two tailored AI systems implemented in 14 public hospitals into four likelihood levels for COVID-19.ResultsBetween March and October 2020, 911 COVID-19 diagnoses were performed in 14 regional hospitals (62.6% were men and 37.4% were women). The average age of the patients diagnosed with COVID-19 was 50.7 years; 59.1% were aged 19 to 59 years. The two AI systems used have different background information for detecting COVID-19. The most common findings were severe pneumonia and bilateral pneumonia with pleural effusions. The role of computed tomography was to find lesions and evaluate the effects of treatment. The sensitivity of AI for detecting COVID-19 was 93%.ConclusionsAI technology could help in developing a screening system for COVID-19 and other respiratory pathologies. It could speed up medical imaging diagnosis at regional hospitals for patients with suspected infection during the COVID-19 pandemic and rationalize scarce RT-PCR and specialized human resources in low-income countries. These results must be contextualized with the local or regional epidemiological profile before widespread implementation.

Introduction: Due to the different epidemiology of COVID-19 in different regions of the world, it is important to know the impact of health variables specific to each country. Objective:To evaluate the role of history of BCG vaccination and recent history of dengue among risk factors for hospitalization of patients with COVID-19.Methods: Observational, cross-sectional study that recorded sociodemographic and clinical variables by means of structured interview in patients diagnosed with COVID-19 in four health institutions in Paraguay (September to December 2020). Logistic regression models evaluated factors associated with outcome.Results: 397 patients were included. The frequency of hospitalization was higher in male patients, age > 40 years, lower income, obesity, hypertension and diabetes mellitus. There was less hospitalization among health personnel and in those who reported bronchial asthma. Male sex (ORa 3.72; arterial hypertension (ORa 2.46;, income (ORa 1.98; 95%CI 1.03-3.80), healthcare worker (ORa 0.20; 95%CI 0.11-0.37) and bronchial asthma (ORa 0.40; 95%CI 0.20-0.82) were influential in the multivariate analysis. In certain models studied by logistic regression, those who reported a history of BCG vaccination were associated with a lower frequency of hospitalization. History of symptomatic dengue fever was not among the relevant variables related to outcome.Conclusions: Among several COVID-19 severity prediction models, BCG vaccination history may be associated with hospitalization rates.