RenalTwin: Diseño y validación de un gemelo digital para el modelado predictivo y la detección temprana de lesión renal aguda en pacientes críticos
DOI:
https://doi.org/10.67166/dygxgh35Palabras clave:
gemelo digital; lesión renal aguda; cuidados intensivos; modelado predictivo; medicina de precisión.Resumen
La lesión renal aguda (LRA) continúa siendo una complicación de alto impacto en las unidades de cuidados intensivos y su reconocimiento puede ocurrir después de que la alteración funcional ya se ha establecido. El objetivo fue diseñar y validar metodológicamente RenalTwin, un gemelo digital orientado al modelado predictivo y a la detección temprana de LRA en pacientes críticos. Se desarrolló un estudio cuantitativo, correlacional-descriptivo y de validación predictiva multicéntrica in silico con una cohorte sintética reproducible de 2.480 perfiles clínicos distribuidos en cuatro centros virtuales. El gemelo integró creatinina basal, diuresis, presión arterial media, lactato, Sequential Organ Failure Assessment (SOFA), soporte vasopresor, ventilación mecánica, sepsis, exposición nefrotóxica, nitrógeno ureico y variables demográficas. La cohorte se dividió en entrenamiento (60 %), calibración (20 %) y prueba (20 %); se estimaron área bajo la curva ROC (AUROC), área bajo la curva precisión-recall (AUPRC), sensibilidad, especificidad, valor predictivo, F1-score, Brier, calibración, t de Student pareada y tamaño del efecto d de Cohen. En prueba, RenalTwin alcanzó AUROC de 0,916 (IC95 % bootstrap aproximado: 0.890-0.941), AUPRC de 0,856 (IC95 %: 0.805-0.901), sensibilidad de 0,814, especificidad de 0,862, F1-score de 0,770 y Brier de 0,106. El adelanto temporal sintético de alerta fue de 14.3 horas respecto al criterio convencional, con diferencia estadísticamente significativa (p<0,001; d=3.42). Se concluye que la arquitectura propuesta muestra discriminación, calibración y utilidad potencial suficientes para justificar una validación prospectiva con datos clínicos reales. Los hallazgos no constituyen evidencia de eficacia clínica y deben interpretarse como una validación metodológica previa a implementación.
Descargas
Referencias
Alfieri, F., Ancona, A., Tripepi, G., et al. (2023). Continuous and early prediction of future moderate and severe acute kidney injury in critically ill patients: Development and multi-centric, multi-national external validation of a machine-learning model. PLOS ONE, 18(7), e0287398. https://doi.org/10.1371/journal.pone.0287398
Baeseman, L., Gunning, S., & Koyner, J. L. (2024). Biomarker enrichment in sepsis-associated acute kidney injury: Finding high-risk patients in the intensive care unit. American Journal of Nephrology, 55(1), 72–85. https://doi.org/10.1159/000534608
Cannon, J. W., Gruen, D. S., Zamora, R., et al. (2024). Digital twin mathematical models suggest individualized hemorrhagic shock resuscitation strategies. Communications Medicine, 4, 113. https://doi.org/10.1038/s43856-024-00535-6
Collins, G. S., Moons, K. G. M., Dhiman, P., Riley, R. D., Beam, A. L., Van Calster, B., et al. (2024). TRIPOD+AI statement: Updated guidance for reporting clinical prediction models that use regression or machine learning methods. BMJ, 385, e078378. https://doi.org/10.1136/bmj-2023-078378
Gottlieb, E. R., Samuel, M., Bonventre, J. V., Celi, L. A., & Mattie, H. (2022). Machine learning for acute kidney injury prediction in the intensive care unit. Advances in Chronic Kidney Disease, 29(5), 431–438. https://doi.org/10.1053/j.ackd.2022.06.005
Havaldar, A. A., Sushmitha, E. A. C., Shrouf, S. B., et al. (2024). Epidemiological study of hospital acquired acute kidney injury in critically ill and its effect on the survival. Scientific Reports, 14, 28129. https://doi.org/10.1038/s41598-024-79533-6
Hu, C., Tan, Q., Zhang, Q., Li, Y., Wang, F., Zou, X., & Peng, Z. (2022). Application of interpretable machine learning for early prediction of prognosis in acute kidney injury. Computational and Structural Biotechnology Journal, 20, 2861–2870. https://doi.org/10.1016/j.csbj.2022.06.003
Huang, C.-T., Wang, T.-J., Kuo, L.-K., Tsai, M.-J., Cia, C.-T., Chiang, D.-H., et al. (2023). Federated machine learning for predicting acute kidney injury in critically ill patients: A multicenter study in Taiwan. Health Information Science and Systems, 11, 48. https://doi.org/10.1007/s13755-023-00248-5
Jiang, W., Zhang, Y., Weng, J., Song, L., Liu, S., Li, X., et al. (2025). Explainable machine learning model for predicting persistent sepsis-associated acute kidney injury: Development and validation study. Journal of Medical Internet Research, 27, e62932. https://doi.org/10.2196/62932
Jiang, X., Hu, Y., Guo, S., Du, C., & Cheng, X. (2022). Prediction of persistent acute kidney injury in postoperative intensive care unit patients using integrated machine learning: A retrospective cohort study. Scientific Reports, 12, 17134. https://doi.org/10.1038/s41598-022-21428-5
Kamel Rahimi, A., Ghadimi, M., van der Vegt, A. H., et al. (2023). Machine learning clinical prediction models for acute kidney injury: The impact of baseline creatinine on prediction efficacy. BMC Medical Informatics and Decision Making, 23, 207. https://doi.org/10.1186/s12911-023-02306-0
Kuruppu Appuhamilage, G. D. K., Hussain, M., Zaman, M., et al. (2025). A health digital twin framework for discrete event simulation based optimised critical care workflows. npj Digital Medicine, 8, 376. https://doi.org/10.1038/s41746-025-01738-4
Lee, K. H., Yoon, D., Lim, H., Lee, K.-B., & Lee, Y. K. (2026). Deep learning models for acute kidney injury prediction: Multi-center external validation and evaluation under simulated continuous monitoring conditions. npj Digital Medicine, 9, 544. https://doi.org/10.1038/s41746-026-02722-2
Lekadir, K., Frangi, A. F., Porras, A. R., Glocker, B., Cintas, C., Langlotz, C. P., et al. (2025). FUTURE-AI: International consensus guideline for trustworthy and deployable artificial intelligence in healthcare. BMJ, 388, e081554. https://doi.org/10.1136/bmj-2024-081554
Li, X., Wang, P., Zhu, Y., Zhao, W., Pan, H., & Wang, D. (2024). Interpretable machine learning model for predicting acute kidney injury in critically ill patients. BMC Medical Informatics and Decision Making, 24, 148. https://doi.org/10.1186/s12911-024-02537-9
Li, X., Wu, R., Zhao, W., et al. (2023). Machine learning algorithm to predict mortality in critically ill patients with sepsis-associated acute kidney injury. Scientific Reports, 13, 5223. https://doi.org/10.1038/s41598-023-32160-z
Luo, X.-Q., Yan, P., Duan, S.-B., Kang, Y.-X., Deng, Y.-H., Liu, Q., Wu, T., & Wu, X. (2022). Development and validation of machine learning models for real-time mortality prediction in critically ill patients with sepsis-associated acute kidney injury. Frontiers in Medicine, 9, 853102. https://doi.org/10.3389/fmed.2022.853102
Moons, K. G. M., Damen, J. A. A., Kaul, T., et al. (2025). PROBAST+AI: An updated quality, risk of bias, and applicability assessment tool for prediction models using regression or artificial intelligence methods. BMJ, 388, e082505. https://doi.org/10.1136/bmj-2024-082505
Ostermann, M., Legrand, M., Meersch, M., et al. (2024). Biomarkers in acute kidney injury. Annals of Intensive Care, 14, 145. https://doi.org/10.1186/s13613-024-01360-9
Persson, I., Grünwald, A., Morvan, L., et al. (2023). A machine learning algorithm predicting acute kidney injury in intensive care unit patients (NAVOY acute kidney injury): Proof-of-concept study. JMIR Formative Research, 7, e45979. https://doi.org/10.2196/45979
Rovati, L., Gary, P. J., Cubro, E., Dong, Y., Kilickaya, O., Schulte, P. J., Zhong, X., Wörster, M., Kelm, D. J., Gajic, O., Niven, A. S., & Lal, A. (2024). Development and usability testing of a patient digital twin for critical care education: A mixed methods study. Frontiers in Medicine, 10, 1336897. https://doi.org/10.3389/fmed.2023.1336897
Stanski, N. L., Rodrigues, C. E., Strader, M., Murray, P. T., Endre, Z. H., & Bagshaw, S. M. (2023). Precision management of acute kidney injury in the intensive care unit: Current state of the art. Intensive Care Medicine, 49(9), 1049–1061. https://doi.org/10.1007/s00134-023-07171-z
Tran, T. T., Yun, G., & Kim, S. (2024). Artificial intelligence and predictive models for early detection of acute kidney injury: Transforming clinical practice. BMC Nephrology, 25, 353. https://doi.org/10.1186/s12882-024-03793-7
Trevena, W., Zhong, X., Lal, A., Rovati, L., Cubro, E., Dong, Y., Schulte, P., & Gajic, O. (2024). Model-driven engineering for digital twins: A graph model-based patient simulation application. Frontiers in Physiology, 15, 1424931. https://doi.org/10.3389/fphys.2024.1424931
Vagliano, I., Hsu, W.-H., & Schut, M. C. (2022). Machine learning, clinical notes and knowledge graphs for early prediction of acute kidney injury in the intensive care. Studies in Health Technology and Informatics, 289, 329–332. https://doi.org/10.3233/SHTI210926
Vasey, B., Nagendran, M., Campbell, B., Clifton, D. A., Collins, G. S., Denaxas, S., et al. (2022). Reporting guideline for the early stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI. BMJ, 377, e070904. https://doi.org/10.1136/bmj-2022-070904
Weaver, L., Shamohammadi, H., Saffaran, S., Tonelli, R., Laviola, M., Laffey, J. G., Camporota, L., Scott, T. E., Hardman, J. G., Clini, E., & Bates, D. G. (2024). Digital twins of acute hypoxemic respiratory failure patients suggest a mechanistic basis for success and failure of noninvasive ventilation. Critical Care Medicine, 52(9), e473–e484. https://doi.org/10.1097/CCM.0000000000006337
Zappalà, S., Alfieri, F., Ancona, A., Taccone, F. S., Maviglia, R., Cauda, V., Finazzi, S., & Dell’Anna, A. M. (2024). Development and external validation of a machine learning model for the prediction of persistent acute kidney injury stage 3 in multi-centric, multi-national intensive care cohorts. Critical Care, 28, 189. https://doi.org/10.1186/s13054-024-04954-8
Zhang, K., Zhou, H.-Y., Baptista-Hon, D. T., Gao, Y., Liu, X., Oermann, E., et al. (2024). Concepts and applications of digital twins in healthcare and medicine. Patterns, 5(8), 101028. https://doi.org/10.1016/j.patter.2024.101028
Zhang, L., Li, M., Wang, C., Zhang, C., & Wu, H. (2025). Prediction of acute kidney injury in intensive care unit patients based on interpretable machine learning. DIGITAL HEALTH, 11, 20552076241311173. https://doi.org/10.1177/20552076241311173
Descargas
Publicado
Número
Sección
Licencia
Derechos de autor 2026 Christian Gustavo Masapanta Salazar (Autor/a)
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial-CompartirIgual 4.0.
