Computed tomography volumetry vs three-dimensional surface scanning for liver graft size estimation using explant liver mass

Figure 1 Comparison of computed tomography volumetry, three-dimensional reconstruction workflow, and corresponding correlation analyses. A: Representative axial computed tomography (CT) image; B: Three-dimensional (3D) liver model reconstructed from CT imaging using semi-automated segmentation and volumetric analysis (AW Server 3.2 EXT 4.0, GE Healthcare; Hepatic VCAR application); C: 3D reconstructed explant model generated from Polycam (photogrammetry-based scanning application) and refined in Blender, representing the surface-based anatomical morphology of the liver specimen; D: Correlation between 3D surface-derived volumetry and liver explant mass; E: Correlation between CT volumetry and explant liver mass. Values are expressed in grams (g) for mass and cubic centimeters (cm³) for volumetric measurements. CT: Computed tomography; 3D: Three-dimensional.

Figure 2 Agreement between imaging-based graft size estimates and explant liver mass. Bland-Altman plots showing the differences between imaging-based graft size estimates and explant liver mass plotted against the mean of the measurements. A: Computed tomography volumetry vs explant liver mass; B: Three-dimensional surface-derived volumetry vs explant liver mass. Dashed lines indicate the mean bias and 95% limits of agreement. Values are expressed in grams (g). CT: Computed tomography; 3D: Three-dimensional.

Figure 3 Absolute error of graft size estimation relative to explant liver mass for computed tomography volumetry and three-dimensional surface scanning. A: Three-dimensional (3D) surface reconstruction of a liver explant; B: Boxplot distribution of absolute percentage error (%) for computed tomography volumetry and 3D surface-derived volumetry; C: 3D reconstruction of the liver from computed tomography imaging used for volumetric analysis. CT: Computed tomography; 3D: Three-dimensional.