CoChat
Sign Up
Cited by 25Machine Learning in HealthcareArtificial Intelligence in HealthcareOrgan Transplantation Techniques and Outcomes

A transformer-based deep learning approach for fairly predicting post-liver transplant risk factors

Open access
Li, Can, Jiang, Xiaoqian, Zhang, Kai·2024·Journal of Biomedical Informatics
DOI 10.1016/j.jbi.2023.104545PMC PMC11619923
Open this paper in CoChat — free account

Abstract

Liver transplantation is a life-saving procedure for patients with end-stage liver disease. There are two main challenges in liver transplant: finding the best matching patient for a donor and ensuring transplant equity among different subpopulations. The current MELD scoring system evaluates a patient's mortality risk if not receiving an organ within 90 days. However, the donor-patient matching should also consider post-transplant risk factors, such as cardiovascular disease, chronic rejection, etc., which are all common complications after transplant. Accurate prediction of these risk scores remains a significant challenge. In this study, we used predictive models to solve the above challenges. Specifically, we proposed a deep learning model to predict multiple risk factors after a liver transplant. By formulating it as a multi-task learning problem, the proposed deep neural network was trained to simultaneously predict the five post-transplant risks and achieve equal good performance by exploiting task-balancing techniques. We also proposed a novel fairness-achieving algorithm to ensure prediction fairness across different subpopulations. We used electronic health records of 160,360 liver transplant patients, including demographic information, clinical variables, and laboratory values, collected from the liver transplant records of the United States from 1987 to 2018. The model's performance was evaluated using various performance metrics such as AUROC and AUPRC. Our experiment results highlighted the success of our multi-task model in achieving task balance while maintaining accuracy. The model significantly reduced the task discrepancy by 39 %. Further application of the fairness-achieving algorithm substantially reduced fairness disparity among all sensitive attributes (gender, age group, and race/ethnicity) in each risk factor. It underlined the potency of integrating fairness considerations into the task-balancing framework, ensuring robust and fair predictions across multiple tasks and diverse demographic groups.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1016/j.jbi.2023.104545

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • In a study predicting post-liver transplant risk factors, a multi-level approach using in- and post-processing techniques significantly enhanced equitable predictions across subgroups including age, gender, and race/ethnicity.

    “one study focused on predicting post-liver transplant risk factors found that employing a multi-level approach with in- and post-processing techniques significantly enhanced equitable predictions across various subgroups, including age, gender, and race/ethnicity [13].”

    From Advancing Health Equity through Multi-Level Fairness in Health Informatics (2026) · cites this paper as [13]

    The Results section reports significantly reduced fairness disparities across gender, age, and race/ethnicity subgroups. However, the Methodology describes task balancing followed by fairness optimization during model training, not a combination of in- and post-processing techniques; the Related works section discusses post-processing as a separate approach. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 29

  1. [1]Nitski O, Azhie A, Qazi-Arisar FA, Wang X, Ma S, Lilly L, Watt KD, Levitsky J, Asrani SK, Lee DS, Rubin BB, Bhat M, Wang B, Long-term mortality risk stratification of liver transplant recipients: real-time application of deep learning algorithms on longitudinal data, Lancet Digit Health 3 (2021) e295– e305. 33858815
  2. [2]Kantidakis G, Putter H, Lancia C, de Boer J, Braat AE, Fiocco M, Survival prediction models since liver transplantation - comparisons between Cox models and machine learning techniques, BMC Med. Res. Methodol. 20 (2020) 277. 33198650 PMC7667810
  3. [3]John PR, Thuluvath PJ, Outcome of liver transplantation in patients with diabetes mellitus: a case-control study, Hepatology 34 (2001) 889– 895. 11679959
  4. [4]Zhang Y, Yang Q, A survey on multi-task learning, IEEE Trans. Knowl. Data Eng. 34 (2022) 5586– 5609. PMC10619966 37915376
  5. [5]Parent B, Caplan AL, Fair is fair: We must re-allocate livers for transplant, BMC Med. Ethics. 18 (2017) 26. 28381305 PMC5382421
  6. [6]Gentry SE, Massie AB, Cheek SW, Lentine KL, Chow EH, Wickliffe CE, Dzebashvili N, Salvalaggio PR, Schnitzler MA, Axelrod DA, Segev DL, Addressing geographic disparities in liver transplantation through redistricting, Am. J. Transplant. 13 (2013) 2052– 2058. 23837931 PMC4674218
  7. [7]Akshat S, Gentry SE, Raghavan S, Heterogeneous donor circles for fair liver transplant allocation, Health Care Manag. Sci. (2022), 10.1007/s10729-022-09602-7. PMC10896798 35854169 · DOI
  8. [8]Dureja P, Lucey MR, Disparities in liver transplantation in the post–model for end-stage liver disease era: Are we there yet?#, Hepatology 50 (2009) 981. 19670420
  9. [9]Mathur AK, Schaubel DE, Gong Q, Guidinger MK, Merion RM, Sex-based disparities in liver transplant rates in the United States, Am. J. Transplant. 11 (2011) 1435– 1443. 21718440 PMC3132137
  10. [10]Bertsimas D, Papalexopoulos T, Trichakis N, Wang Y, Hirose R, Vagefi PA, Balancing efficiency and fairness in liver transplant access: Tradeoff curves for the assessment of organ distribution policies, Transplantation 104 (2020) 981– 987. 31644494
  11. [11]Lau L, Kankanige Y, Rubinstein B, Jones R, Christophi C, Muralidharan V, Bailey J, Machine-learning algorithms predict graft failure after liver transplantation, Transplantation 101 (2017) e125– e132. 27941428 PMC7228574
  12. [12]Liu C-L, Soong R-S, Lee W-C, Jiang G-W, Lin Y-C, Predicting short-term survival after liver transplantation using machine learning, Sci. Sci. Rep. 10 (2020). PMC7101323 32221367
  13. [13]Ferrarese A, Sartori G, Orrù G, Frigo AC, Pelizzaro F, Burra P, Senzolo M, Machine learning in liver transplantation: a tool for some unsolved questions? Transpl. Int. 34 (2021) 398– 411. 33428298
  14. [15]Mehrabi N, Morstatter F, Saxena N, Lerman K, Galstyan A, A Survey on bias and fairness in machine learning, ACM Comput. Surv. 54 (2021) 1– 35.
  15. [16]Calmon FP, Wei D, Ramamurthy KN, Varshney KR, Optimized data pre-processing for discrimination prevention, ArXiv [Stat.ML]. (2017). https://proceedings.neurips.cc/paper/2017/hash/9a49a25d845a483fae4be7e341368e36-Abstract.html (accessed December 12, 2022).
  16. [17]Pleiss G, Raghavan M, Wu F, Kleinberg J, Weinberger KQ, On fairness and calibration, Adv. Neural Inf. Process. Syst. 30 (2017), in: https://proceedings.neurips.cc/paper/2017/hash/b8b9c74ac526fffbeb2d39ab038d1cd7-Abstract.html.
  17. [18]Wan M, Zha D, Liu N, Zou N, In-processing modeling techniques for machine learning fairness: A survey, ACM Trans. Knowl. Discov. Data 17 (2023) 1– 27.
  18. [19]Ding S, Tang R, Zha D, Zou N, Zhang K, Jiang X, Hu X, Fairly predicting graft failure in liver transplant for organ assigning, AMIA Annu. Symp. Proc. 2022 (2022) 415– 424. 37128420 PMC10148275
  19. [20]Li C, Ding S, Zou N, Hu X, Jiang X, Zhang K, Multi-task learning with dynamic re-weighting to achieve fairness in healthcare predictive modeling, J. Biomed. Inform. 143 (2023), 104399. 37211197 PMC10665114
  20. [21]Li B, Shi X, Gao H, Jinag X, Zhang K, Harmanci AO, Malin B, Enhancing fairness in disease prediction by optimizing multiple domain adversarial networks, BioRxiv (2023), 10.1101/2023.08.04.551906. PMC12124548 40445951 · DOI
  21. [22]Yu T, Kumar S, Gupta A, Levine S, Hausman K, Finn C, Gradient surgery for multi-task learning, Adv. Neural Inf. Process. Syst. 33 (2020) 5824– 5836.
  22. [23]Thinh PTD, Hung HM, Yang H-J, Kim S-H, Lee G-S, Emotion Recognition with Incomplete Labels Using Modified Multi-task Learning Technique, ArXiv [Cs. CV]. (2021). http://arxiv.org/abs/2107.04192. · arXiv
  23. [25]Navon A, Shamsian A, Achituve I, Maron H, Kawaguchi K, Chechik G, Fetaya E, Multi-Task Learning as a Bargaining Game, ArXiv [Cs.LG]. (2022). http://arxiv.org/abs/2202.01017. · arXiv
  24. [26]Ding S, Tan Q, Chang C-Y, Zou N, Zhang K, Hoot NR, Jiang X, Hu X, Multi-Task Learning for Post-transplant Cause of Death Analysis: A Case Study on Liver Transplant, ArXiv [Cs.LG]. (2023). http://arxiv.org/abs/2304.00012. PMC10785876 38222347 · arXiv
  25. [27]Khurana D, Koli A, Khatter K, Singh S, Natural language processing: state of the art, current trends and challenges, Multimed. Tools Appl. 82 (2023) 3713– 3744. 35855771 PMC9281254
  26. [28]Huang X, Khetan A, Cvitkovic M, Karnin Z, TabTransformer: Tabular Data Modeling Using Contextual Embeddings, ArXiv [Cs.LG]. (2020). http://arxiv.org/abs/2012.06678. · arXiv
  27. [29]Ba JL, Kiros JR, Hinton GE, Layer Normalization, ArXiv [Stat.ML]. (2016). http://arxiv.org/abs/1607.06450. · arXiv
  28. [30]Lin T-Y, Goyal P, Girshick R, He K, Dollár P, Focal Loss for Dense Object Detection, ArXiv [Cs.CV]. (2017). http://arxiv.org/abs/1708.02002. 30040631 · arXiv
  29. [33]Azur MJ, Stuart EA, Frangakis C, Leaf PJ, Multiple imputation by chained equations: what is it and how does it work? Int. J. Methods Psychiatr. Res. 20 (2011) 40– 49. 21499542 PMC3074241

Cited by

Sorted by citations

Authors
Li, C., Jiang, X., & Zhang, K.
Affiliations
The University of Texas Health Science Center at Houston (US)
Corresponding author
Zhang, Kai
ORCID
Venue
Journal of Biomedical Informatics
Publisher
Elsevier BV
Published
2024-01
Volume
149
Pages
104545
ISSN
1532-0464 (ISSN-L)
DOI
10.1016/j.jbi.2023.104545
PMC
PMC11619923
OpenAlex
W4388830665
Cited by
25
References
1
Citations by year
2026: 4 · 2025: 13 · 2024: 6 · 2023: 2
FWCI
2.36
Type
journal article
Language
en
Open access
Licence
Locations
  • Journal of Biomedical Informatics (publishedVersion)
  • PubMed (publishedVersion)
  • PubMed Central (acceptedVersion)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • Digital Commons-TMC (Texas Medical Center) (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
  • National Science Foundation (2124789, RR180012, U01TR002062)
  • U.S. Department of Health and Human Services (R01LM013712, R01AG066749, U01TR002062)
  • Cancer Prevention and Research Institute of Texas (R01AG066749, RR180012, U01TR002062)
  • University of Texas Health Science Center at Houston (RR180012)
  • National Institutes of Health (U01TR002062, RR180012)
  • Health Resources and Services Administration (HHSH250201900001C)
  • National Institute on Aging (R01AG066749)
  • U.S. National Library of Medicine (U24LM013755, R01LM013712)
  • National Center for Advancing Translational Sciences
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Machine Learning in Healthcare
  • Health Sciences › Health Professions › Health Information Management › Artificial Intelligence in Healthcare
  • Health Sciences › Medicine › Surgery › Organ Transplantation Techniques and Outcomes
Keywords
liver transplantation, multi-task learning, transformer models, algorithmic fairness, task balancing, transplant equity, MELD score, cardiovascular disease, chronic rejection, electronic health records, demographic disparities
MeSH terms
Deep Learning, Humans, Risk Factors, Tissue Donors, United States, Liver Transplantation, Neural Networks, Computer

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review