*Artificial Intelligence in Surgery, Surgical Subspecialties, and Related Disciplines DOI: http://dx.doi.org/10.5772/intechopen.112691*

patient characteristics, such as age, sex, severity of disease, hemodynamic measurements, and other variables, could be used to predict waitlist mortality and posttransplant outcomes [39]. These programs could be used to improve patient outcomes more broadly through a more objective management of organ transplant waitlists and recipient match optimization. ML algorithms may also be used in the future in more direct applications to transplant surgery. For instance, in liver transplantation, graft-weight-to-recipient-body-weight (GW/RW) ratios <0.8% are associated with an increased risk of complications such as small-for-size syndrome [40]. Consequently, the estimation of graft weight in living donors is important for limiting adverse outcomes associated with graft size mismatch. Studies have been conducted on the potential use of ML models trained on donor age, sex, body mass index, CT scans, and other data to estimate the donor graft weight [40]. These models have the potential to greatly enhance the precision of graft weight estimation, improving outcomes of liver transplantation. Additionally, experiences learned from hepatic transplantation may be suitable for adoption across other areas of organ transplantation (e.g., kidney, pancreas, heart, and lung), similarly reducing various potentially



