Role Of Artificial Intelligence In Improving Access To COVID-19 Diagnosis During Pandemic

ABSTRACT

IntroductionThe evolution of advances in informatics, technology in medicine, and artificial intelligence (AI) offers opportunities to enhance health care during the coronavirus disease 2019 (COVID-19) pandemic. The challenge for biomedical engineers is to implement these developments in clinical practice to improve global health. Populations living in low-income countries do not have access to specialist care and quality diagnostic services for COVID-19. Therefore, an AI system based on a telemedicine platform for diagnosing COVID-19 could help mitigate the lack of highly trained radiologists at regional hospitals and serve as a triage system for rationalizing the use of reverse transcription polymerase chain reaction (RT-PCR) testing and other health resources in low-income countries. Thus, the utility of an AI system for diagnosing COVID-19 in Paraguay was investigated.MethodsThis is a descriptive multicenter observational feasibility study of an AI tool for the rapid detection of COVID-19 in chest computed tomography (CT) images of patients with respiratory difficulties who attended public hospitals across the country.ResultsBetween March 2020 and August 2021, 3,514 rapid diagnostic tests were carried out on patients with respiratory disorders to rule out COVID-19 in 14 hospitals nationwide. The average age of the patients was 48.6 years (52.8% were men);the most common age ranges were 27 to 59 years, followed by older than 60 years and 19 to 26 years. The most frequent findings on the CT images 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. Overall, there was 93 percent agreement and 7 percent discordance between the AI system and the RT-PCR test results. Compared with RT-PCR testing, the AI system had a sensitivity of 93 percent and a specificity of 80 percent.ConclusionsParaguay has an AI-based telemedicine screening system for the rapid detection of COVID-19 that uses chest CT images of patients with respiratory conditions.