Fig 1.
(a) normal lung with a regular pleural line, A-lines (green), and lung sliding; (b) infected lung with pleural line broken (red) and fused B-lines (green), (c) subpleural consolidation. Images were taken from the POCUS database [46].
Fig 2.
Flowchart for systematic diagnosis of lung diseases by means of LUS.
(DVT—Deep Venous Thrombosis).
Table 1.
CNN configurations considered for COVID-19 identification in LUS imagery.
Fig 3.
Examples of images obtained from LUS videos after pre-processing.
Healthy (first row), pneumonia (second row), and COVID-19 (third row) LUS image classes.
Fig 4.
5 × 5-fold cross-validation results: Confusion matrices.
Table 2.
Overall performance results of the evaluated models (95% C.I.).
Fig 5.
5 × 5-fold cross-validation results: ACC, BACC, and AUC-ROC ACC, BACC, and AUC-ROC scores boxplots.
Box extends from the Q1 to Q3 quartile values of the data, with a line at the median and a triangle at the mean.
Fig 6.
5 × 5-fold cross-validation results: Mean ROC curves and AUC scores (95% C.I.).
Fig 7.
Post-hoc pairwise comparison using the Wilcoxon signed-rank test with Holm correction.
Table 3.
Overall COVID-19 vs pneumonia performance results (95% C.I.).
Table 4.
Overall COVID-19 vs non-COVID-19 performance results (95% C.I.).