Use of artificial intelligence as an innovative donor-recipient matching model for liver transplantation: Results from a multicenter Spanish study

Briceño, Javier; Cruz-Ramírez, M.; Prieto, M.; Navasa, Miguel; Urbina, J. Ortiz de; Ortí, Rafael; Gómez-Bravo, Miguel-Ángel; Otero, Alejandra; Varó, Evaristo; Tomé, Santiago; Clemente, G.; Bañares, Rafael; Bárcena, Rafael; Cuervas-Mons, Valentín; Solórzano, G; Vinaixa, Carmen; Rubín, Ángel; Colmenero, Jordi; Valdivieso, A.; Ciria, Rubén; Hervás-Martı́nez, César; Mata, Manuel de la

doi:10.1016/j.jhep.2014.05.039

Cited by 101 publications

(94 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…While, the MELD score, that is the foundation of liver allocation on transplant waiting lists[27-29] is a poor predictor of outcome after transplant[30]. A number of other complex models detailing interactions between donor and recipient risk profiles have been developed to predict graft and patient survival after liver transplantation[30-39]. But none consider DCD in isolation and it is well recognized, that the DCD liver is a different type of graft in comparison to DBD, and DBD predictors of outcome have not been found to be applicable to DCD[40].…”

Section: Discussionmentioning

confidence: 99%

Developing a donation after cardiac death risk index for adult and pediatric liver transplantation

Khorsandi¹,

Giorgakis²,

Vilca‐Melendez³

et al. 2017

WJT

View full text Add to dashboard Cite

AIMTo identify objective predictive factors for donor after cardiac death (DCD) graft loss and using those factors, develop a donor recipient stratification risk predictive model that could be used to calculate a DCD risk index (DCD-RI) to help in prospective decision making on organ use.METHODSThe model included objective data from a single institute DCD database (2005-2013, n = 261). Univariate survival analysis was followed by adjusted Cox-regressional hazard model. Covariates selected via univariate regression were added to the model via forward selection, significance level P = 0.3. The warm ischemic threshold was clinically set at 30 min. Points were given to each predictor in proportion to their hazard ratio. Using this model, the DCD-RI was calculated. The cohort was stratified to predict graft loss risk and respective graft survival calculated.RESULTSDCD graft survival predictors were primary indication for transplant (P = 0.066), retransplantation (P = 0.176), MELD > 25 (P = 0.05), cold ischemia > 10 h (P = 0.292) and donor hepatectomy time > 60 min (P = 0.028). According to the calculated DCD-RI score three risk classes could be defined of low (DCD-RI < 1), standard (DCD-RI 2-4) and high risk (DCD-RI > 5) with a 5 years graft survival of 86%, 78% and 34%, respectively.CONCLUSIONThe DCD-RI score independently predicted graft loss (P < 0.001) and the DCD-RI class predicted graft survival (P < 0.001).

show abstract

Section: Discussionmentioning

confidence: 99%

Developing a donation after cardiac death risk index for adult and pediatric liver transplantation

Khorsandi¹,

Giorgakis²,

Vilca‐Melendez³

et al. 2017

WJT

View full text Add to dashboard Cite

show abstract

“…The combination of demographic features, the underlying liver disease, tumour burden, histological characteristics and serum biomarkers by using novel multivariate approaches allowing to manage an increased amount of information such as machine learning classifiers or artificial neural networks, which have already proven their utility in other LT scenarios [58] , might be the key for a safe expansion of the Milan criteria.…”

Section: Namentioning

confidence: 99%

Strategies to improve outcome of patients with hepatocellular carcinoma receiving a liver transplantation

Guerrero-Misas

Rodríguez‐Perálvarez

Mata

2015

WJH

View full text Add to dashboard Cite

Liver transplantation is the only therapeutic option which allows to treat both, the hepatocellular carcinoma and the underlying liver disease. Indeed, liver transplantation is considered the standard of care for a subset of patients with cirrhosis and hepatocellular carcinoma.However, tumour recurrence rates are as high as 20%, and once the recurrence is established the therapeutic options are scarce and with little impact on prognosis. Strategies to minimize tumour recurrence and thus to improve outcome may be classified into 3 groups: (1) An adequate selection of candidates for liver transplantation by using the Milan criteria; (2) An optimized management within waiting list including prioritization of patients at high risk of tumour progression, and the implementation of bridging therapies, particularly when the expected length within the waiting list is longer than 6 mo; and (3) Tailored immunosuppression comprising reduced exposure to calcineurin inhibitors, particularly early after liver transplantation, and the addition of mammalian target of rapamycin inhibitors. In the present manuscript the available scientific evidence supporting these strategies is comprehensively reviewed, and future directions are provided for novel research approaches, which may contribute to the final target: to cure more patients with hepatocellular carcinoma and with an improved long term outcome. Core tip: Liver transplantation is the only therapeutic option which allows to treat both, the hepatocellular carcinoma and the underlying liver disease. However, tumour recurrence rates are 15%-20% with a very poor prognosis. Strategies to minimize tumour recurrence and thus to improve outcome are focused in a careful selection of candidates for liver transplantation, an optimized management within waiting list and a tailored immunosuppression. The available scientific evidence supporting these strategies is reviewed, and future directions are provided for novel research approaches, which may contribute to the final target: to cure more REVIEW Submit a

show abstract

“…Deep neural networks (DNNs) have been a tremendous breakthrough in ML, enabling machines to learn patterns of data by modeling them through a combination of simple nonlinear elementary operations. Neural networks have been applied to predict 3‐month graft survival and assist with donor‐recipient matching for patients with end‐stage liver disease as well as predicting the presence of liver disease from imaging . This can be further extended into convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which handle local structures and sequential data consecutively .…”

Section: Tools and Applications To Liver Diseasementioning

confidence: 99%

“…In efforts to address this, ML has been used to improve equity of donor‐recipient matching by using rule‐based systems using artificial neural networks (ANNs) . Using a compilation of clinical data from 1,003 LT recipients and their donors, as well as data from the retrieval and pretransplant process, ANNs were used to generate separate algorithms for prediction of 3‐month graft survival . When compared to several current scoring systems that use either isolated donor/recipient scores or combined donor/recipient factors, the ANNs more accurately predicted 3‐month graft survival (AUC, 0.81) and graft loss (AUC, 0.82).…”

Section: In Ltmentioning

confidence: 99%

“…Neural networks have been applied to predict 3-month graft survival and assist with donor-recipient matching for patients with end-stage liver disease as well as predicting the presence of liver disease from imaging. (20,21) This can be further extended into convolutional neural networks (CNNs) and recurrent neural networks (RNNs), which handle local…”

Section: Tools and Applications To Liver Diseasementioning

confidence: 99%

See 1 more Smart Citation

Applying Machine Learning in Liver Disease and Transplantation: A Comprehensive Review

et al. 2020

View full text Add to dashboard Cite

Machine learning (ML) utilizes artificial intelligence to generate predictive models efficiently and more effectively than conventional methods through detection of hidden patterns within large data sets. With this in mind, there are several areas within hepatology where these methods can be applied. In this review, we examine the literature pertaining to machine learning in hepatology and liver transplant medicine. We provide an overview of the strengths and limitations of ML tools and their potential applications to both clinical and molecular data in hepatology. ML has been applied to various types of data in liver disease research, including clinical, demographic, molecular, radiological, and pathological data. We anticipate that use of ML tools to generate predictive algorithms will change the face of clinical practice in hepatology and transplantation. This review will provide readers with the opportunity to learn about the ML tools available and potential applications to questions of interest in hepatology.

show abstract

Use of artificial intelligence as an innovative donor-recipient matching model for liver transplantation: Results from a multicenter Spanish study

Cited by 101 publications

References 38 publications

Developing a donation after cardiac death risk index for adult and pediatric liver transplantation

Developing a donation after cardiac death risk index for adult and pediatric liver transplantation

Strategies to improve outcome of patients with hepatocellular carcinoma receiving a liver transplantation

Applying Machine Learning in Liver Disease and Transplantation: A Comprehensive Review

Contact Info

Product

Resources

About