Fig. 9
From: A deep learning framework for automated detection and quantitative assessment of liver trauma
Liver trauma and parenchymal segmentation results on two patients who were determined to be outliers based on Bland–Altman analysis. (a) Axial contrast-enhanced CT image from Patient Y in Fig. 8. Patient Y has the largest trauma region in our dataset. (b) Axial contrast-enhanced CT image from Patient Z in Fig. 8. Patient Z’s CT image is distorted by linear streak artifact, which leads to a large false positive region of segmented trauma. In (a) and (b), the first column corresponds to the original CT image; the second column corresponds to the ground truth annotations; and the third column corresponds to the automated segmentation results. In both ground truth annotations and the segmentation results, the green line shows the liver contour while the red line marks the contour of trauma regions