Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Lung segmentations in 12-zone lung POCUS.

The chest was divided into eight zones for the ventral wall and four zones for the dorsal chest wall, divided by the posterior axillary line (PAL). The parasternal line (PSL) and paravertebral line (PVL) to two hemithorax. The ventral chest wall was segmented by the midline and anterior axillary lines (AAL) and further divided into upper and lower zones. The dorsal chest wall was divided into left-right and upper-lower zones at the height of the tip of the scapula.

More »

Fig 1 Expand

Fig 2.

Ultrasound B-lines.

B-lines were automatically analyzed using the real-time deep learning-based application, which counted the number of B-lines (indicated by the white lines in the upper panel) in the movie, and recorded the maximum number of B-lines. Three or more B-lines in each zone were considered abnormal (red zone in lower panel). Patients with one or more abnormal zones were considered positive for AI-POCUS.

More »

Fig 2 Expand

Fig 3.

Patient inclusion chart.

Only 4 patients (8.9%) of cases were excluded and 56 patients (41 cases and 15 controls) were included in the final analysis.

More »

Fig 3 Expand

Table 1.

Patient characteristic.

More »

Table 1 Expand

Fig 4.

Distribution of pneumonia.

The X-axis represents the zone of the lung (R1, right anterior superior zone; R2, right anterior inferior zone; R3, right lateral superior zone; R4, right lateral inferior zone; R5, right superior dorsal zone; R6, right inferior dorsal zone; Ls are left zones corresponding to Rs), and the Y-axis represents the number of patients with pneumonia in each zone.

More »

Fig 4 Expand

Fig 5.

Accuracy of AI-POCUS versus CT.

Each panel is a confusion matrix showing the concordance between the AI-POCUS results and CT-validated pneumonia diagnosis.

More »

Fig 5 Expand