Transforming liver transplant allocation with artificial intelligence and machine learning: a systematic review.

Area of Science:

• Medical Informatics
• Artificial Intelligence in Medicine
• Transplant Surgery

Background:

• Organ allocation decisions are guided by urgency, utility, and benefit principles.
• Liver transplant (LT) allocation currently relies heavily on urgency-based models.
• Existing models may not fully capture the complexities of optimal organ distribution.

Purpose of the Study:

• To identify and analyze data elements used in Machine Learning (ML) and Artificial Intelligence (AI) for LT.
• To examine data sources and their focus on urgency, utility, or benefit in LT allocation.
• To assess the current state of ML/AI application in optimizing LT decisions.

Main Methods:

• Comprehensive literature search in Ovid Medline and Scopus (2002-June 2023).
• Inclusion of quantitative studies employing ML/AI for LT candidates, donors, or recipients.
• Data extraction and eligibility assessment following PRISMA guidelines by two independent reviewers.

Main Results:

• Twenty papers were included, categorized into five themes.
• ML/AI models primarily focus on donor-recipient matching and predicting post-LT survival.
• Studies highlight ML/AI potential but note limitations in transplant-related benefit models and urgency assessments compared to MELD.

Conclusions:

• AI and ML offer significant potential to enhance LT allocation and patient outcomes.
• Advancements exist, but improved urgency and transplant-related benefit models are needed.
• Future research should focus on model interpretability and generalizability for better organ allocation and survival prediction.