人工智能在肾移植领域中的应用

Application of artificial intelligence in the field of kidney transplantation

摘要

肾移植临床诊疗涉及大量高维、非线性和多模态的复杂数据，传统统计学方法难以充分识别其潜在规律，而人工智能（AI）技术具备更强的数据整合与分析能力，尤其擅长处理复杂数据。本文综述了 AI 在肾移植领域中的应用，涵盖了移植前阶段深度学习模型在供肾质量量化评估中的作用、虚拟活检系统的价值和供-受者匹配优化策略，以及移植后阶段 iBox 系统对移植物存活率的动态预测、AI 模型在排斥反应无创监测中的优势和免疫抑制药个体化剂量预测的应用成果。尽管医疗 AI 临床应用仍受限于模型可解释性不足、泛化能力存疑和伦理与法律风险，但联邦学习与数字孪生等技术有望解决数据隐私与仿真预测难题，进一步推动肾移植的精准化发展。

Abstract

The clinical diagnosis and treatment of kidney transplantation involve a large amount of high-dimensional, nonlinear and multimodal complex data. Traditional statistical methods make it difficult to fully identify the underlying patterns. However, artificial intelligence (AI) technology has stronger capabilities in data integration and analysis, and is particularly adept at handling complex data. This article reviews the application of AI in the field of kidney transplantation, covering the role of deep learning models in the quantitative assessment of donor kidney quality in the pre-transplantation stage, the value of virtual biopsy systems and the optimization strategies for donor-recipient matching, as well as the dynamic prediction of graft survival rate by the iBox system in the post-transplantation stage, the advantages of AI models in non-invasive monitoring of rejection and the application results of individualized dose prediction of immunosuppressive drugs. Although the clinical application of medical AI is still limited by insufficient model interpretability, questionable generalization ability, and ethical and legal risks, technologies such as federated learning and digital twins are expected to solve the problems of data privacy and simulation prediction, and further promote the precise development of kidney transplantation.

关键词

肾移植 / 人工智能 / 机器学习 / 深度学习 / iBox 系统 / 移植物排斥反应 / 全视野数字切片 / 肾脏供者特征指数

Key words

Kidney transplantation / Artificial intelligence / Machine learning / Deep learning / iBox system / Graft rejection / Whole slide image / Kidney donor profile index

参考文献

[1] RAJKOMAR A, DEAN J, KOHANE I. Machine learning in medicine[J]. N Engl J Med, 2019, 380(14): 1347-1358. DOI: 10.1056/nejmra1814259.

[2] HUNTER J, HOSGOOD S, MOERS C, et al. Improving outcomes in kidney transplantation through advances in donor organ perfusion[J]. Nat Rev Nephrol, 2025, 21(12): 818-832. DOI: 10.1038/s41581-025-00993-8.

[3] 中华医学会器官移植学分会, 中国医疗保健国际交流促进会肾脏移植学分会. 中国肾脏移植受者免疫抑制治疗指南（2023 版）[J]. 中华器官移植杂志, 2024, 45(10): 645-663. DOI: 10.3760/cma.j.cn421203-20240418-00092.

[4] Branch of Organ Transplantation of Chinese Medical Association, Branch of Kidney Transplantation of China International Exchange and Promotive Association for Medical and Health Care. Chinese clinical practice guidelines for immunosuppressive therapy in kidney transplant recipients (2023 edition)[J]. Chin J Organ Transplant, 2024, 45(10): 645-663. DOI: 10.3760/cma.j.cn421203-20240418-00092.

[5] MIZERA J, PONDEL M, KEPINSKA M, et al. Advancements in artificial intelligence for kidney transplantology: a comprehensive review of current applications and predictive models[J]. J Clin Med, 2025, 14(3): 975. DOI: 10.3390/jcm14030975.

[6] HE Y J, LIU P L, WEI T, et al. Artificial intelligence in kidney transplantation: a 30-year bibliometric analysis of research trends, innovations, and future directions[J]. Ren Fail, 2025, 47(1): 2458754. DOI: 10.1080/0886022X.2025.2458754.

[7] LAI W, MIN J, PENG L. Will AI replace physicians? a review of artificial intelligence applications in kidney transplantation[J]. Organ Med, 2025, 2(1): 15-30. DOI: 10.1002/orm2.70001.

[8] WANG J, YAN J, ZHANG T, et al. Predicting adverse outcomes in kidney transplant recipients using an interpretable model based on shear-wave elastography[J]. Kidney Dis, 2025, 11(1): 469-481. DOI: 10.1159/000546396.

[9] KOTSIFA E, MAVROEIDIS V K. Present and future applications of artificial intelligence in kidney transplantation[J]. J Clin Med, 2024, 13(19): 5939. DOI: 10.3390/jcm13195939.

[10] MATSUO Y, LECUN Y, SAHANI M, et al. Deep learning, reinforcement learning, and world models[J]. Neural Netw, 2022, 152: 267-275. DOI: 10.1016/j.neunet.2022.03.037.

[11] YE Y, XIA L, YANG S, et al. Deep learning-enabled classification of kidney allograft rejection on whole slide histopathologic images[J]. Front Immunol, 2024, 15: 1438247. DOI: 10.3389/fimmu.2024.1438247.

[12] HART A, LENTINE K L, SMITH J M, et al. OPTN/SRTR 2019 annual data report: kidney[J]. Am J Transplant, 2021, 21(Suppl 2): 21-137. DOI: 10.1111/ajt.16502.

[13] MARSH J N, LIU T C, WILSON P C, et al. Development and validation of a deep learning model to quantify glomerulosclerosis in kidney biopsy specimens[J]. JAMA Netw Open, 2021, 4(1): e2030939. DOI: 10.1001/jamanetworkopen.2020.30939.

[14] YI Z, XI C, MENON M C, et al. A large-scale retrospective study enabled deep-learning based pathological assessment of frozen procurement kidney biopsies to predict graft loss and guide organ utilization[J]. Kidney Int, 2024, 105(2): 281-292. DOI: 10.1016/j.kint.2023.09.031.

[15] YOO D, DIVARD G, RAYNAUD M, et al. A machine learning-driven virtual biopsy system for kidney transplant patients[J]. Nat Commun, 2024, 15(1): 554. DOI: 10.1038/s41467-023-44595-z.

[16] COEMANS M, NAESENS M. Continuous donor-recipient age-matching in the Eurotransplant region: outcomes, parameters and prediction[J]. Am J Transplant, 2025, 25(4): 883-884. DOI: 10.1016/j.ajt.2025.01.011.

[17] GOTLIEB N, AZHIE A, SHARMA D, et al. The promise of machine learning applications in solid organ transplantation[J]. NPJ Digit Med, 2022, 5(1): 89. DOI: 10.1038/s41746-022-00637-2.

[18] PLACONA A M, MARTINEZ C, MCGEHEE H, et al. Can donor narratives yield insights? a natural language processing proof of concept to facilitate kidney allocation[J]. Am J Transplant, 2020, 20(4): 1095-1104. DOI: 10.1111/ajt.15705.

[19] HAMOUDA E, EL-METWALLY S, TAREK M. Ant lion optimization algorithm for kidney exchanges[J]. PLoS One, 2018, 13(5): e0196707. DOI: 10.1371/journal.pone.0196707.

[20] KAWAKITA S, BEAUMONT J L, JUCAUD V, et al. Personalized prediction of delayed graft function for recipients of deceased donor kidney transplants with machine learning[J]. Sci Rep, 2020, 10(1): 18409. DOI: 10.1038/s41598-020-75473-z.

[21] JEN K Y, ALBAHRA S, YEN F, et al. Automated en masse machine learning model generation shows comparable performance as classic regression models for predicting delayed graft function in renal allografts[J]. Transplantation, 2021, 105(12): 2646-2654. DOI: 10.1097/TP.0000000000003640.

[22] MARK E, GOLDMAN D, GURBAXANI B, et al. Using machine learning and an ensemble of methods to predict kidney transplant survival[J]. PLoS One, 2019, 14(1): e0209068. DOI: 10.1371/journal.pone.0209068.

[23] SENANAYAKE S, KULARATNA S, HEALY H, et al. Development and validation of a risk index to predict kidney graft survival: the kidney transplant risk index[J]. BMC Med Res Methodol, 2021, 21(1): 127. DOI: 10.1186/s12874-021-01319-5.

[24] MULUGETA G, ZEWOTIR T, TEGEGNE A S, et al. Classification of imbalanced data using machine learning algorithms to predict the risk of renal graft failures in Ethiopia[J]. BMC Med Inform Decis Mak, 2023, 23(1): 98. DOI: 10.1186/s12911-023-02185-5.

[25] LOUPY A, AUBERT O, ORANDI B J, et al. Prediction system for risk of allograft loss in patients receiving kidney transplants: international derivation and validation study[J]. BMJ, 2019, 366: l4923. DOI: 10.1136/bmj.l4923.

[26] KLEIN A, LOUPY A, STEGALL M, et al. Qualifying a novel clinical trial endpoint (iBOX) predictive of long-term kidney transplant outcomes[J]. Transpl Int, 2023, 36: 11951. DOI: 10.3389/ti.2023.11951.

[27] HOGAN J, GEORGE R, HAYES C, et al. Donor-derived cell-free DNA as a noninvasive biomarker of kidney allograft rejection in pediatric kidney transplantation[J]. Transplantation, 2025, 109(9): 1520-1525. DOI: 10.1097/TP.0000000000005403.

[28] KUEHT M L, DONGUR L P, CUSICK M, et al. The current state of donor-derived cell-free DNA use in allograft monitoring in kidney transplantation[J]. J Pers Med, 2022, 12(10): 1700. DOI: 10.3390/jpm12101700.

[29] AUBERT O, URSULE-DUFAIT C, BROUSSE R, et al. Cell-free DNA for the detection of kidney allograft rejection[J]. Nat Med, 2024, 30(8): 2320-2327. DOI: 10.1038/s41591-024-03087-3.

[30] VELUT Y, POULET G, BERSEZ T, et al. Epigenetic signatures on plasma cell-free DNA to detect kidney allograft rejection in a non-invasive way: development of a 10-plex digital PCR assay[J]. Biomark Res, 2025, 13(1): 118. DOI: 10.1186/s40364-025-00834-7.

[31] HALLORAN P F, REEVE J, MADILL-THOMSEN K S, et al. The trifecta study: comparing plasma levels of donor-derived cell-free DNA with the molecular phenotype of kidney transplant biopsies[J]. J Am Soc Nephrol, 2022, 33(2): 387-400. DOI: 10.1681/ASN.2021091191.

[32] TASCA P, VAN DEN BERG B M, RABELINK T J, et al. Application of spatial-omics to the classification of kidney biopsy samples in transplantation[J]. Nat Rev Nephrol, 2024, 20(11): 755-766. DOI: 10.1038/s41581-024-00861-x.

[33] 闵建亮, 陈国栋. 机器学习在肾移植受者他克莫司剂量预测中的应用[J]. 器官移植, 2025, 16(6): 921-930. DOI: 10.12464/j.issn.1674-7445.2025215.

[34] MIN J L, CHEN G D . Application of machine learning in tacrolimus dose prediction for kidney transplant recipients[J]. Organ Transplant, 2025, 16(6): 921-930. DOI: 10.12464/j.issn.1674-7445.2025215.

[35] PILCH N A, BOWMAN L J, TABER D J. Immunosuppression trends in solid organ transplantation: the future of individualization, monitoring, and management[J]. Pharmacotherapy, 2021, 41(1): 119-131. DOI: 10.1002/phar.2481.

[36] WOILLARD J B, LABRIFFE M, DEBORD J, et al. Tacrolimus exposure prediction using machine learning[J]. Clin Pharmacol Ther, 2021, 110(2): 361-369. DOI: 10.1002/cpt.2123.

[37] GOLDSTEIN B A, MOHOTTIGE D, BESSIAS S, et al. Enhancing clinical decision support in nephrology: addressing algorithmic bias through artificial intelligence governance[J]. Am J Kidney Dis, 2024, 84(6): 780-786. DOI: 10.1053/j.ajkd.2024.04.008.

[38] QI X, WANG S, FANG C, et al. Machine learning and SHAP value interpretation for predicting comorbidity of cardiovascular disease and cancer with dietary antioxidants[J]. Redox Biol, 2025, 79: 103470. DOI: 10.1016/j.redox.2024.103470.

[39] DICKINSON Q, MEYER J G. Positional SHAP (PoSHAP) for interpretation of machine learning models trained from biological sequences[J]. PLoS Comput Biol, 2022, 18(1): e1009736. DOI: 10.1371/journal.pcbi.1009736.

[40] ACHILONU O, OBAIDO G, OGBUOKIRI B, et al. A machine learning approach towards assessing consistency and reproducibility: an application to graft survival across three kidney transplantation eras[J]. Front Digit Health, 2024, 6: 1427845. DOI: 10.3389/fdgth.2024.1427845.

[41] 张艾一, 余中光. 基于国内外政策分析的医疗人工智能伦理风险治理研究[J]. 中国医学伦理学, 2024, 37(9): 1061-1067. DOI: 10.12026/j.issn.1001-8565.2024.09.07.

[42] ZHANG A Y, YU Z G . Research on ethical risk governance of medical artificial intelligence based on domestic and foreign policy analysis[J]. Chin Med Ethics, 2024, 37(9): 1061-1067. DOI: 10.12026/j.issn.1001-8565.2024.09.07.

[43] ZHANG L, RICHTER L R, KIM T, et al. Evaluating and improving the performance and racial fairness of algorithms for GFR estimation[PP/OL]. medRxiv(2024-04-23) [2026-01-03]. https://ieeexplore.ieee.org/document/1050436.

[44] 杨金铭, 王纳, 胡业勋, 等 . 人工智能医疗中的法律风险防范[J]. 四川大学学报（医学版）, 2025, 56(1): 143-148. DOI: 10.12182/20250160301.

[45] YANG J M, WANG N, HU Y X, et al. Legal risk assessment and prevention in artificial intelligence-assisted health care[J]. J Sichuan Univ Med Sci, 2025, 56(1): 143-148. DOI: 10.12182/20250160301.

[46] RAMPERSAD C, BAU J T, ORCHANIAN-CHEFF A, et al. iBox as a predictor of kidney allograft failure: a systematic review[J]. Am J Kidney Dis, 2025, 86(2): 276-285. DOI: 10.1053/j.ajkd.2024.12.007.

[47] XU J, GLICKSBERG B S, SU C, et al. Federated learning for healthcare informatics[J]. J Healthc Inform Res, 2021, 5(1): 1-19. DOI: 10.1007/s41666-020-00082-4.

[48] KHOSHFEKR RUDSARI H, TSENG B, ZHU H, et al. Digital twins in healthcare: a comprehensive review and future directions[J]. Front Digit Health, 2025, 7: 1633539. DOI: 10.3389/fdgth.2025.1633539.

基金资助

广东省自然科学基金面上项目(2024A1515010473)