Clin Surg | Volume 3, Issue 1 | Research Article | Open Access

Is Artificial Intelligence the Most Reliable Way to Predict Mortality After Liver Transplantation?

Marcos Bruna Esteban1*, Eva Montalvá2, Antonio J. Serrano-López3, Joan Vila-Francés3, Javier Maupoey2 and Juan Vila J2

1Department of General and Digestive Surgery, General University Hospital of Valencia, Spain
2University and Polytechnic Hospital of La Fe, Spain
3Department of Data Analysis Laboratory, School of Engineering at the University of Valencia, Spain

*Correspondance to: Marcos Bruna Esteban 

Fulltext PDF

Abstract

Introduction: Graft allocation in Liver Transplantation (LT) should be based on the greatestsurvival benefit to the patients awaiting transplantation. This study developed a predictive model to determine recipient mortality 1 year after LT.Materials and
Methods:
We developed Artificial Neural Network (ANN) and Logistic Regression (LR) models and compared their results with the Balance of Risk (BAR), Survival Outcomes Following Transplantation (SOFT), and Model for End-stage Liver Disease (MELD) scores. The Development Group used to create the predictive models included 1235 valid cases, while 200 consecutive transplant patients since January 2009 were included in the Generalization Group for internal validation.Results: The area under the curve (AUC) of the ANN model (0.82) was higher than that of the LR model (0.68). For the Generalization Group, the MELD, SOFT, and BAR scores had AUCs of 0.56, 0.57, and 0.62, respectively. The ANN model had a significantly higher AUC than that of each score (MELD, p=0.005; SOFT, p=0.009; BAR, p=0.02).Conclusion: ANN model was superior to the LR model and other scales currently used to predict mortality during the first year after LT and to match a particular graft with potential recipients.

Keywords:

Artificial Intelligence; Liver Transplantation; Recipient Survival; Waitlist Management

Citation:

Esteban MB, Montalv� E, Serrano-L�pez AJ, Vila-Franc�s J, Maupoey Jm Juan Vila J. Is Artificial Intelligence the Most Reliable Way to Predict Mortality After Liver Transplantation? Clin Surg. 2018; 3: 2122.

Subscribe to Our Newsletter