Artificial Intelligence in Liver Transplantation: A Systematic Review

Area of Science:

• Transplantation Medicine
• Medical Informatics
• Computational Biology

Background:

• Artificial intelligence (AI) offers advanced computational methods for integrating complex datasets in transplantation.
• Liver transplantation, with its high-dimensional data and critical decisions, is an ideal area for AI applications.

Purpose of the Study:

• To systematically review AI applications in liver transplantation.
• To analyze data modalities, algorithms, outcomes, validation, and performance metrics of AI in this field.

Main Methods:

• Systematic review of 1045 records from MEDLINE, Scopus, and Cochrane Library.
• Inclusion of 65 studies focusing on AI in liver transplantation.

Main Results:

• Laboratory and clinical data were most common inputs; deep learning predominated.
• AI models showed high predictive performance, often exceeding traditional scores for outcomes like graft survival.
• Imaging and genomic AI demonstrated strong segmentation and discriminative capabilities.

Conclusions:

• AI shows significant promise for enhancing clinical decision-making in liver transplantation.
• Methodological limitations include data heterogeneity, validation issues, and interpretability challenges.
• Robust prospective validation and standardized frameworks are crucial for AI adoption in liver transplantation.