Using Artificial Intelligence for Predicting Survival of Individual Grafts in Liver Transplantation: A Systematic Review

Abstract

Background:The demand for liver transplantation far outstrips the supply of deceased donor organs, and so listing and allocation decisions aim to maximise utility. Most existing methods for predicting transplant outcomes utilise basic methods such as regression modellingnewer artificial intelligence techniques have the potential to improve predictive accuracy. Aims:To systematically review studies predicting graft outcomes following deceased liver transplantation using Artificial Intelligence (AI) techniques and comparing these to linear regression and standard predictive modelling (donor risk index, DRI; Model for end-stage liver disease, MELD; survival outcome following liver transplantation, SOFT).Methods: A systematic review was performed. PubMed, Cochrane, MEDLINE, Science Direct, Springer Link, Elsevier, and reference lists were analysed for appropriate inclusion.Results: A total of 52 papers were reviewed for inclusion. Of these papers, 9 met the inclusion criteria, reporting outcomes from 18,771 liver transplants. Artificial neural networks (ANN) were the most commonly utilised methodology, being reported in 7 studies. Two studies directly compared Machine Learning (ML) techniques to liver scoring modalities (i.e. DRI, SOFT, BAR). Both of these studies showed better prediction of individual organ survival with the optimal ANN model reporting AUC ROC 0.82 compared with BAR: 0.62 and SOFT: 0.57; and the other ANN model showing an AUC ROC: 0.84 compared to DRI: 0.68 and SOFT: 0.64. Conclusion:AI techniques can provide high accuracy in predicting graft survival based on donors and recipient variables. AI approaches need to be integrated into current organ allocation methodologies to ultimately create better individual patient outcomes and will personalise decisions to organ acceptance.

Augmenting the Transplant Team With Artificial Intelligence: Toward Meaningful AI Use in Solid Organ Transplant

“…We further note that not all studies compare the AI tools they have developed to the status quo "non-AI tool" (e.g. MELD or SOFT scores) (12). Even so, evaluating accuracy, sensitivity, specificity, the Receiver Operator Curve (ROC) AUC, etc., will only provide insight to how well an AI makes decisions against retrospective data.…”

Abdominal Organ Transplantation: Noteworthy Literature in 2020

“…They found 2 studies that directly compared machine learning techniques with liver scoring modalities (ie, DRI, SOFT, and BAR) for liver graft outcome. 55 One of the studies was done by Briceño et al who showed the machine learning AUROC of 0.82 was superior to compared with a BAR of 0.62 and a SOFT of 0.57. 56 The other research by Lau et al showed that a random forest and artificial neural network model revealed an AUROC of 0.84 compared with a DRI of 0.68 and SOFT of 0.64.…”

“…Cardiovascular adverse effects are common with terlipressin and need to be monitored strictly. • Artificial intelligence and risk prediction: Wingfield et al 55 A systematic review of artificial intelligence computing techniques used in liver transplantation to predict individual patient graft survival. They found 2 studies suggesting the superiority of machine learning techniques to standard liver scoring modalities for liver graft outcome.…”

“…However, Wingfield and associates cautioned that artificial intelligence’s high accuracy might come at the cost of losing explainability (to patients and clinicians) on how the technology works. 55…”

AI in colonoscopy and beyond: On the cusp of clinical implementation?

“…These endpoints must be measured in prospective trials. Research has highlighted promise of AI across gastroenterology, from analysing liver ultrasound images 5 to prognosticating in hepatocellular carcinoma, 6 and from personalising pancreatic cancer management 7 to predicting liver transplantation survival 8 . However, the impact of incorporating such algorithms into clinical workflows has not yet been robustly assessed 9 .…”