Machine learning in liver transplantation: a tool for some unsolved questions?
Alberto Ferrarese1, Giuseppe Sartori2, Graziella Orrù3
1Multivisceral Transplant Unit, Department of Surgery, Oncology and Gastroenterology, Padua University Hospital, Padua, Italy.
Summary
Machine learning enhances medical diagnosis and prognosis. In liver transplantation, these models predict survival, optimize donor matching, and forecast outcomes, improving patient care.
Area of Science:
- Medical Informatics
- Computational Medicine
- Transplantation Surgery
Background:
- Machine learning (ML) offers advanced tools for improving diagnostic and prognostic accuracy in medicine.
- Solid organ transplantation involves complex, nonlinear relationships influencing patient prognosis.
- ML models are increasingly applied to transplantation to address these complexities.
Purpose of the Study:
- To review the application of machine learning in liver transplantation.
- To highlight the strengths and limitations of ML in this field.
- To discuss future perspectives for ML in liver transplant outcomes.
Main Methods:
- Narrative review of existing literature on machine learning in liver transplantation.
- Analysis of ML model applications for pretransplant survival prediction.
- Evaluation of ML for donor-recipient matching and postoperative outcome forecasting.
Main Results:
- ML models have been developed to predict pretransplant survival in cirrhosis patients.
- Applications include assessing optimal donor-to-recipient matching.
- ML aids in foreseeing postoperative complications and overall outcomes.
Conclusions:
- Machine learning shows significant potential in enhancing liver transplantation.
- Understanding ML's strengths and pitfalls is crucial for its effective implementation.
- Future research should focus on refining ML applications for improved liver transplant care.
