Korean J Transplant.  2023 Nov;37(Suppl 1):S311. 10.4285/ATW2023.F-9210.

Introducing free software for liver volumetry

• Kim KW
• Lee S
• Kwon HJ
• Lee J

• KMID: 2551940
• DOI: http://doi.org/10.4285/ATW2023.F-9210

Abstract

We introduce free downloadable software for semi-automated hepatic volumetry running on 5 mm computed tomography (CT) images. The precise URL will be open at presentation. The overall procedure for liver volumetry is illustrated in figure. First, the initial liver outline is detected through the sequential applications of seeded region growing onto level-set speed images. Malladis level-set method is adopted for liver segmentation based on the initially detected liver contour. The level-set speed function is formed by combining the curvature speed component, inversely proportional to the curvature of the zero level-set curve, with the propagation speed component. The zero level-set curve moves proportionally to the propagation speed. The curvature speed term regularizes the zero level-set curve by smoothing the high curvature region. Then, a rolling ball algorithm and the removal of false positives are performed to enhance the liver boundary more accurately. Secondly, liver vessel is identified by interactive thresholding with optimization of a lower limit to include the vessels identified on axial CT scans as much as possible while excluding the parenchyma. Finally, the operator can repeatedly define a resection line dividing the liver into the target volume (e.g., right or left hemiliver) and the remnant. When resection lines are drawn on at least two slices, all resection lines on slices in between can be computed utilizing the interpolation technique, and the control points of cubic Hermite spline curve are generated in the slice-by-slice manner. And the operator can modify control points for the resection lines in details in order that the target liver can be accurately divided from the rest part of the liver.