AI in Liver Transplantation Shows Promise but Rarely Leaves the Retrospective Zone

Artificial intelligence is being explored across nearly every stage of liver transplantation, from matching donors to predicting long-term complications, but a comprehensive scoping review suggests the field remains far from routine clinical use. The review, covering MEDLINE and Embase literature from 2000 to September 2026, identified 196 primary articles reporting on clinical AI applications in liver transplantation.

The distribution of research was uneven. Long-term post-operative complications dominated the literature, accounting for 45.9% of studies, followed by the pre-operative period before transplant at 21.9% and short-term post-operative complications at 16.8%. Organ retrieval attracted 9.7% of studies, intra-operative issues just 3.6%, and natural language processing a mere 2.0%.

The review’s most striking findings concern study design. Some 95.9% of the identified studies were retrospective, meaning they analysed existing patient records rather than testing AI tools in real time. Only 12.8% used direct multi-centre data, with the remainder relying on national registries or single-centre datasets. Just 13.8% of studies externally validated their AI tools, a key step in demonstrating that a model performs reliably on data it was not trained on.

Prospective evaluation was rarer still: only one study in the entire review assessed its tool through a randomized controlled trial, the gold standard for establishing clinical benefit before deployment.

The authors conclude that progress towards clinically deployable AI tools will require harmonised, higher-granularity datasets, consistent reporting standards, and prospective evaluation embedded within real-world transplant workflows. Until then, AI in liver transplantation remains largely a retrospective research exercise rather than a validated component of perioperative care.

Subject of Research: Clinical applications of artificial intelligence in liver transplantation

Article Title: The applications of artificial intelligence in liver transplantation: A comprehensive scoping review

Article References: Borakati, A., Agilinko, J., Wu, J., Hunter, S., Alhaq, A., Froghi, F., Teo, J., Bernal, W., Callaghan, C., Cortes-Cerisuelo, M., & Wallace, D. (2026). The applications of artificial intelligence in liver transplantation: A comprehensive scoping review. PLOS One, 21(10), e0361120. https://doi.org/10.1371/journal.pone.0361120

Image Credits: AI Generated

DOI: 10.1371/journal.pone.0361120

Keywords: artificial intelligence, liver transplantation, scoping review, machine learning, retrospective studies, external validation, randomized controlled trial, clinical deployment, organ retrieval, post-operative complications, natural language processing, transplant workflows