RESUMEN
Chest CT scan is currently used to assess the extent of lung involvement in patients with the coronavirus disease 2019 (COVID-19). The aim of this study was to evaluate the diagnostic performance of lung ultrasound in the diagnosis of COVID-19 pulmonary manifestations in comparison to CT scan. Thirty-three symptomatic patients with suspected COVID-19 pneumonia were evaluated by lung ultrasound and then, at a short interval, chest CT scan. In the anterior chest, each hemithorax was divided into four areas. In the posterior chest, eight zones similar to the anterior part were examined. The axillary areas were also divided into upper and lower zones (20 zones were determined per patient). Mean age of the patients was 58.66 years. The sensitivity (95% CI) and specificity (95% CI) of lung ultrasound for the diagnosis of parenchymal lesions were 90.5% (69.6-98.8%) and 50% (21.1-78.9%), respectively. In the evaluation of pleural lesions, the sensitivity (95% CI) and specificity (95% CI) of lung ultrasound were 100% (71.5-100%) and 22.7% (7.8-45.4%), respectively. Owing to the high sensitivity of ultrasound in identifying lung lesions in patients with COVID-19 pneumonia, it can be recommended to use lung ultrasound as a tool for initial screening of patients with high clinical suspicion for SARS-CoV-2 infection during the pandemic. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s13337-021-00736-w.
RESUMEN
[This corrects the article DOI: 10.1371/journal.pone.0250952.].
RESUMEN
The development of medical assisting tools based on artificial intelligence advances is essential in the global fight against COVID-19 outbreak and the future of medical systems. In this study, we introduce ai-corona, a radiologist-assistant deep learning framework for COVID-19 infection diagnosis using chest CT scans. Our framework incorporates an EfficientNetB3-based feature extractor. We employed three datasets; the CC-CCII set, the MasihDaneshvari Hospital (MDH) cohort, and the MosMedData cohort. Overall, these datasets constitute 7184 scans from 5693 subjects and include the COVID-19, non-COVID abnormal (NCA), common pneumonia (CP), non-pneumonia, and Normal classes. We evaluate ai-corona on test sets from the CC-CCII set, MDH cohort, and the entirety of the MosMedData cohort, for which it gained AUC scores of 0.997, 0.989, and 0.954, respectively. Our results indicates ai-corona outperforms all the alternative models. Lastly, our framework's diagnosis capabilities were evaluated as assistant to several experts. Accordingly, We observed an increase in both speed and accuracy of expert diagnosis when incorporating ai-corona's assistance.