Fig. 6From: Automated quantitative assessment of pediatric blunt hepatic trauma by deep learning-based CT volumetrySegmented livers and trauma regions along with ground-truth labels in five pediatric patients with different AAST grades. a AAST grade I case with the automated LPDI 3.16% (the reference LPDI is 2.25%). b AAST grade II case with the automated LPDI 3.02% (the reference LPDI is 2.94%). c AAST grade III case with the automated LPDI 6.11% (the reference LPDI is 6.60%). d AAST grade IV case with the automated LPDI 22.82% (the reference LPDI is 24.11%). e AAST grade V case with the automated LPDI 37.66% (the reference LPDI is 38.74%). The first column corresponds to the unlabeled CT image; the second column corresponds to the ground-truth labels; and the third column corresponds to the automated segmentation results. The blue line shows the liver contour while the red line represents the contour of trauma regionsBack to article page