Estimation Of Right Lobe Graft Weight For Living Donor Liver Transplantation Using Deep Learning-based Fully Automatic Computed Tomographic Volumetry

This study aimed to propose a deep learning-based fully automatic technique for division of the liver into the left and right lobes for right lobe graft weight estimation and evaluate its performance. 248 patients (171 males and 77 females; mean age: 50.0 ± 12.3 years; age range: 18-79 years) with healthy livers from Jeonbuk National University hospital from January 2009 to December 2021 were included in the study. The technique uses a commercial software solution which applies a deep learning algorithm for fully automated segmentation of the liver. After segmentation errors are corrected, the resection plane for the right lobe graft is defined based on the Cantlie line by manually drawing two dividing lines.