Long-term mortality risk stratification of liver transplant recipients: real-time application of deep learning algorithms on longitudinal data

Abstract

Background Survival of liver transplant recipients beyond 1 year since transplantation is compromised by an increased risk of cancer, cardiovascular events, infection, and graft failure. Few clinical tools are available to identify patients at risk of these complications, which would flag them for screening tests and potentially life-saving interventions. In this retrospective analysis, we aimed to assess the ability of deep learning algorithms of longitudinal data from two prospective cohorts to predict complications resulting in death after liver transplantation over multiple timeframes, compared with logistic regression models. MethodsIn this machine learning analysis, model development was done on a set of 42 146 liver transplant recipients (mean age 48•6 years [SD 17•3]; 17 196 [40•8%] women) from the Scientific Registry of Transplant Recipients (SRTR) in the USA. Transferability of the model was further evaluated by fine-tuning on a dataset from the University Health Network (UHN) in Canada (n=3269; mean age 52•5 years [11•1]; 1079 [33•0%] women). The primary outcome was cause of death, as recorded in the databases, due to cardiovascular causes, infection, graft failure, or cancer, within 1 year and 5 years of each follow-up examination after transplantation. We compared the performance of four deep learning models against logistic regression, assessing performance using the area under the receiver operating characteristic curve (AUROC). Findings In both datasets, deep learning models outperformed logistic regression, with the Transformer model achieving the highest AUROCs in both datasets (p<0•0001). The AUROC for the Transformer model across all outcomes in the SRTR dataset was 0•804 (99% CI 0•795-0•854) for 1-year predictions and 0•733 (0•729-0•769) for 5-year predictions. In the UHN dataset, the AUROC for the top-performing deep learning model was 0•807 (0•795-0•842) for 1-year predictions and 0•722 (0•705-0•764) for 5-year predictions. AUROCs ranged from 0•695 (0•680-0•713) for prediction of death from infection within 5 years to 0•859 (0•847-0•871) for prediction of death by graft failure within 1 year.Interpretation Deep learning algorithms can incorporate longitudinal information to continuously predict long-term outcomes after liver transplantation, outperforming logistic regression models. Physicians could use these algorithms at routine follow-up visits to identify liver transplant recipients at risk for adverse outcomes and prevent these complications by modifying management based on ranked features.

A Bibliometric Analysis of the Landscape of Pediatric Liver Transplantation

“…Countries/regions in East Asia seem to have fewer international collaborations than those in North America and Europe. The intellectual base of most studies is still from disciplines concerning biomedicine, but many experts are devoted to and appealing for interdisciplinary studies, such as liver transplantation combined with materials science ( 30 ), computer science ( 31 , 32 ), sociology, and ethics ( 6 , 33 ). This possibly hints that because of the lack of interdisciplinary studies, attempts to combine PLT with multiple disciplines may be promising.…”

The digital transformation of hepatology: The patient is logged in

“…In both data sets, DL models outperformed conventional logistic regression, with AUROCs of 0.80 for predictions at 1 year and 0.72–0.73 for predictions at 5 years. [ 55 ]…”

Clinical liver transplant tolerance: Recent topics

“…However, as the period gets longer, the long‐term adverse events of IS have been recognized, such as infections, malignancies, cardiovascular disease, metabolic disorders, and renal and other complications 3 . To avoid such long‐term adverse events of organ transplantation, “immunologic tolerance”, ie the maintenance of normal organ function without IS, has emerged as the ultimate treatment goal 3‐7 . The liver is well known as the most tolerogenic among transplantable organs, 8‐10 so the possibility of new clinical applications is most anticipated in this arena.…”

“…3 To avoid such long-term adverse events of organ transplantation, "immunologic tolerance", ie the maintenance of normal organ function without IS, has emerged as the ultimate treatment goal. [3][4][5][6][7] The liver is well known as the most tolerogenic among transplantable organs, 8-10 so the possibility of new clinical applications is most anticipated in this arena.…”

“…3 To avoid such long-term adverse events of organ transplantation, "immunologic tolerance", ie the maintenance of normal organ function without IS, has emerged as the ultimate treatment goal. [3][4][5][6][7] The liver is well known as the most tolerogenic among transplantable organs, 8-10 so the possibility of new clinical applications is most anticipated in this arena.Numerous animal studies have shown a definite procedure to induce immunologic tolerance in transplantation, but so far, the only elective clinical trials have been for kidney and LT. [11][12][13]…”