Statistical methods versus machine learning techniques for donor-recipient matching in liver transplantation

Abstract

Donor-Recipient (D-R) matching is one of the main challenges to be fulfilled nowadays. Due to the increasing number of recipients and the small amount of donors in liver transplantation, the allocation method is crucial. In this paper, to establish a fair comparison, the United Network for Organ Sharing database was used with 4 different end-points (3 months, and 1, 2 and 5 years), with a total of 39, 189 D-R pairs and 28 donor and recipient variables. Modelling techniques were divided into two groups: 1) classical statistical methods, including Logistic Regression (LR) and Naïve Bayes (NB), and 2) standard machine learning techniques, including Multilayer Perceptron (MLP), Random Forest (RF), Gradient Boosting (GB) or Support Vector Machines (SVM), among others. The methods were compared with standard scores, MELD, SOFT and BAR. For the 5-years end-point, LR (AUC = 0.654) outperformed several machine learning techniques, such as MLP (AUC = 0.599), GB (AUC = 0.600), SVM (AUC = 0.624) or RF (AUC = 0.644), among others. Moreover, LR also outperformed standard scores. The same pattern was reproduced for the others 3 end-points. Complex machine learning methods were not able to improve the performance of liver allocation, probably due to the implicit limitations associated to the collection process of the database.

CONUT Score Predicts Early Morbidity After Liver Transplantation: A Collaborative Study

“…Many patients who were once deemed too frail are now considered for LT (1). However, due to the necessity to fulfill the gap between offer and demand of liver grafts, increased utilization of extended-criteria donors has led to more risky donor-to-recipient matches (2). These challenging matches contribute to post-operative morbidity and poor long-term outcomes (3).…”

A novel method to ensure the security of the shared medical data using smart contracts: Organ transplantation sample

“…These data are included in sensitive data due to they contain the EHRs 1–4 . Among the current problems waiting to be solved, which are donor consent processed with traditional methods, priority order in recipient organ waiting list, priority order in emergency organ waiting list, and recipient‐donor matching problem 5–8 . In addition, it is necessary to be fair in ordering the organ waiting lists.…”

“…In addition, the permissioned chain is the most useful way for sharing sensitive and secure medical data for reasons of P2P network connection, decentralized ecosystem, up‐to‐date consensus protocols, the immutability of data. Hyperledger permissioned/private blockchain environment is more suitable for sharing sensitive and secure medical data 2–14 …”

“…[1][2][3][4] Among the current problems waiting to be solved, which are donor consent processed with traditional methods, priority order in recipient organ waiting list, priority order in emergency organ waiting list, and recipient-donor matching problem. [5][6][7][8] In addition, it is necessary to be fair in ordering the organ waiting lists. Developed priority determination criteria are used to ensure fairness in organ waiting lists.…”

“…Hyperledger permissioned/private blockchain environment is more suitable for sharing sensitive and secure medical data. [2][3][4][5][6][7][8][9][10][11][12][13][14] HLF is a project that provides an environment for developing open-source blockchain supported by the Linux Foundation. It is a distributed ledger platform for running chaincode (smart contract), and evidenced technologies.…”