Predicting mortality in patients with chronic kidney disease on hemodialysis: artificial intelligence versus traditional predictive models
Keywords:
ENFERMEDAD RENAL CRÓNICA, MORTALIDAD, HEMODIÁLISIS, INTELIGENCIA ARTIFICIAL, APRENDIZAJE AUTOMÁTICO.Abstract
Background: traditional predictive models for mortality prediction in patients with chronic kidney disease have numerous limitations that limit their implementation in clinical practice. The application of artificial intelligence algorithms could contribute to improving the accuracy of predictions.
Objective: describe the prediction of mortality in patients with chronic kidney disease on hemodialysis using traditional predictive models and artificial intelligence algorithms.
Methods: a literature review was conducted on mortality prediction in patients with chronic kidney disease on hemodialysis. The following resources were used: PubMed, PubMed Central, SciELO, Web of Science, Scopus, Ebsco, Clinical Key, as well as Google Scholar. The search strategies were [(chronic kidney disease OR renal insufficiency OR end-stage renal disease) AND (hemodialysis) AND (mortality) AND (predictive models) AND (artificial intelligence OR machine learning)].
Results: despite a large amount of literature on prediction methodology, the methods used in many investigations that feature traditional predictive models do not meet standards and the quality of reporting of methods and results is poor. Artificial intelligence algorithms have the ability to analyze large volumes of clinical and biomedical data, identify non-linear relationships, and predict clinical outcomes with unprecedent accuracy, allowing for more informed and personalized therapeutic decisions.
Conclusions: artificial intelligence algorithms have numerous benefits that, if applied in the design of new predictive models, can overcome the limitations of traditional predictive models.
Downloads
References
Bello A, Okpechi I, Osman M, Cho Y, Htay H, Jha V, et al. Epidemiology of haemodialysis outcomes. Nature Reviews Nephrology [revista en Internet]. 2022 [citado 18 de febrero 2025]; 18(1): 378-395. Disponible en: https://doi.org/10.1038/s41581-022-00542-7.
Lim DK, Boyd JH, Thomas E, Chakera A, Tippaya S, Irish A, et al. Prediction models used in the progression of chronic kidney disease: A scoping review. PLoS ONE [revista en internet]. 2022 [citado 18 de febrero 2025]; 17(7). Disponible en: https://doi.org/10.1371/journal.pone.0271619.
Bai Q, Tang W. Artificial intelligence in peritoneal dialysis: general overview. Renal Failure [revista en Internet]. 2022 [citado 18 de febrero 2025]; 44(1): 682-687. Disponible en: https://doi.org/10.1080/0886022X.2022.2064304.
Bai Q, Su C, Tang W, Li Y. Machine learning to predict end stage kidney disease in chronic kidney disease. Scientific Reports [revista en internet]. 2022 [citado 18 de febrero 2025]; 12(1): 8377. Disponible en: https://doi.org/10.1038/s41598-022-12316-z.
Kim J, Steingroever J, Lee K, Oh J, Choi M, Lee J. Clinical Interventions and All-Cause Mortality of Patients with Chronic Kidney Disease: An Umbrella Systematic Review of Meta-Analyses. J. Clin. Med [revista en internet]. 2020 [citado 18 de febrero 2025]; 9(2): 394. Disponible en: https://doi.org/10.3390/jcm9020394
Park S, Park BS, Lee YJ, Kim H, Park JH, Ko J, et al. Artificial intelligence with kidney disease A scoping review with bibliometric analysis, PRISMA-ScR. Medicine [revista en internet]. 2021 [citado 18 Feb 2023]; 100(1): 14. Disponible en: http://dx.doi.org/10.1097/MD.0000000000025422.
Ye Z, An S, Gao Y, Xie E, Zhao X, Guo Z, et al. The prediction of in-hospital mortality in chronic kidney disease patients with coronary artery disease using machine learning models. European Journal of Medical Research [revista en internet]. 2023 [citado 27 de julio 2024]; 28(1): 33. Disponible en: https://doi.org/10.1186/s40001-023-00995-x.
Escalona-González SO, Caballero-Mota Y, Rodríguez-Alvarez Y, León-Acebo M, González-Milán ZC, Ricardo-Paez B, et al. Diseño de una escala predictiva de mortalidad en pacientes con enfermedad renal crónica. Revista Cubana de Medicina Militar [revista en internet]. 2024 [citado 18 de febrero 2025]; 53(1). Disponible en: https://revmedmilitar.sld.cu/index.php/mil/article/view/17622 .
Noordzij M, van Diepen M, Caskey FC. Relative risk versus absolute risk: one cannot be interpreted without the other. Nephrology Dialysis Transplantation [revista en internet]. 2017 [citado 18 de febrero 2025]; 32(2): 13-18. Disponible en: https://doi.org/10.1093/ndt/gfw465.
Marsk A, Fluck N, Prescott GJ. Looking to the future: predicting renal replacemente outcomes in a large community cohort with chronic kidney disease. Nephrology Dialysis Transplantation [revista en internet]. 2015 [citado 18 de febrero 2025]; 30(9): 1507-1517. Disponible en: https://doi.org/10.1093/ndt/gfv089.
Riphagen IJ, Kleefstra N, Drion I. Comparison of methods for renal risk prediction in patients with type 2 diabetes (ZODIAC-36). Plos One [revista en internet]. 2015 [citado 18 de febrero 2025]; 10(3): e0120477. Disponible en: https://doi.org/10.1371/journal.pone.0120477.
Nusinovici S, Tham Y, Yu M, Ting D, Li J. Logistic regression was as good as machine learning for predicting major chronic diseases. Journal of clinical epidemiology [revista en internet]. 2020 [citado 18 de febrero 2025]; 122(1): 56-69. Disponible en: https://doi.org/10.1016/j.jclinepi.2020.03.002.
Lewis RM, Battey HS. On inference in high-dimensional logistic regression models with separated data. Biometrika [revista en internet]. 2023 [citado 18 de febrero 2025]; 65(1): 1-24. Disponible en: https://doi.org/10.1093/biomet/asad065.
Desai R, Wang S, Vaduganathan, Evers T. Comparison of machine learning methods with traditional models for use of administrative claims with elctronic medical records to predict heart failure outcomes. JAMA [revista en internet]. 2020 [citado 18 de febrero 2023]; 3(1). Disponible en: https://doi.org/10.1001/jamanetworkopen.2019.18962.
Escalona-González SO, González-Milán ZC, Ricardo-Paez B. Mortality prediction in chronic kidney disease using artificial intelligence algorithms. Medicine in a Virtual Age [revista en internet]. 2022 [citado 18 de febrero 2025]. Disponible en: https://conferences.nature.com/event/0b87c6fb-30cf-495a-9da0-88b2a9bf0681/summary.
Battineni G, Sagaro GG, Chinatalapudi N, Amenta F. Applications of machine learning predictive models in the chronic disease diagnosis. Journal of personalized medicine [revista en Internet]. 2020 [citado 18 de febrero 2025]; 10(2): 21. Disponible en: https://doi.org/10.3390/jpm10020021.
Sarma K, Harmon S, Snford T, Roth H, Xu Z, Tetreault J. Federated learning improves site performance in multicenter deep learning without data sharing. Journal of the American Medical Informatics Association [revista en internet]. 2021 [citado 18 de febrero 2025]; 28(6): 1259-1264. Disponible en: https://doi.org/10.1093/jamia/ocaa341.
Shyu C, Putra K, Chen H, Tsai Y, Tozammel K. A systematic review of federated learning in the healthcare area: From the perspective of data properties and applications. Applied Sciences [revista en internet]. 2021 [citado 18 de febrero 2025]; 23(11): 11191. Disponible en: https://doi.org/10.3390/app11231191.
Andaur Navarro C, Damen J, van Smeden M, Takada T. Systematic review identifies the design and methodological conduct of studies on machine learning-based prediction models. Journal of clinical epidemiology [revista en Internet]. 2023 [citado 18 de febrero 2025]; 154(1): 8-22. Disponible en: https://doi.org/10.1016/j.clinepi.2022.11.015.
Kumar Y, Koul A, Singla R, Ijaz M. Artificial intelligence in disease diagnosis: a systematic literature review, synthesizing framework and future research agenda. Journal of ambient intelligence and humanized computing [revista en internet]. 2023 [citado 18 de febrero 2025]; 14(7): 8459-8456. Disponible en: https://doi.org/10.1007/s12652-021-03612-z.
Rodríguez-Alvarez Y, Caballero-Mota Y, García-Lorenzo MM, Escalona-González SO. Aplicaciones de la inteligencia artificial para la solución de problemas reales en el sector de la salud. En: Rodríguez-Méndez V, Fernández-Ramírez N, editores. Aplicaciones de la inteligencia artificial en la solución de problemas reales. Camagüey: Ediciones Universidad de Camagüey; 2024. p. 62-108.
Escalona-González SO, Caballero-Mota Y, Rodríguez-Alvarez Y, León-Acebo M, González-Milán ZC, Ricardo-Paez B, et al. Red neuronal artificial para la predicción de mortalidad de pacientes con enfermedad renal crónica. Revista Cubana de Medicina Militar [revista en internet]. 2024 [citado 27 de julio 2024]; 53(3). Disponible en: https://revmedmilitar.sld.cu/index.php/mil/article/view/38408.
Díez-Sanmartín C, Sarasa Cabezuelo A, Belmonte AA. A new approach to predicting mortality in dialysis patients using sociodemographic features based on artificial intelligence. Artificial Intelligence in Medicine [revista en internet]. 2023 [citado 27 de julio 2024]; 136(1): 102478. Disponible en: https://doi.org/10.1016/j.artmed.2022.102478.
Simeri A, Pezzi G, Arena R, Papalia G, Szili-Torok T, Greco R, et al. Artificial intelligence in chronic kidney diseases: methodology and potential applications. International Urology and Nephrology. [revista en internet]. 2025 [citado 10 de enero 2025]; 57(1): 159-168. Disponible en: https://doi.org/10.1007/s11255024-04165-8.
Subramanian M, Wojtusciszyn A, Favre L, Boughorbel S, Shan J, Letaief K, et al. Precision medicine in the era of artificial intelligence: implications in chronic disease management. Journal of Translational Medicine. [revista en internet]. 2020 [citado 10 de enero 2025]; 18(1): 472. Disponible en: https://doi.org/10.1186/s12967-020-02658-5.
Downloads
Additional Files
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Sergio Orlando Escalona González, Zoraida Caridad González-Milán
This work is licensed under a Creative Commons Attribution 4.0 International License.
This journal provides free and immediate access to its content under the principle that making research freely available to the public supports a greater exchange of global knowledge. This means that the authors transfer the copyright to the journal, so that copies and distribution of the contents can be made by any means, as long as the authors' acknowledgment is maintained. These terms are a reflection that the journal assumes copyright under Creative Commons licenses, specifically under a Creative Commons Attribution 4.0 International license.
Registration and submission of articles to the journal is free. The processing, including review, editing and publication, is completely free of charge.