Predicción de readmisión hospitalaria a 30 días mediante arquitectura CNN-LSTM sobre datos clínicos tabulares y series temporales de signos vitales
DOI:
https://doi.org/10.55204/scc.v5i2.e105Palabras clave:
Readmisión hospitalaria, aprendizaje profundo, CNN-LSTM, series temporales clínicas, MIMIC-IV, signos vitales, predicción clínicaResumen
La readmisión hospitalaria no planificada en un periodo de 30 días constituye un indicador crítico de calidad asistencial y un desafío económico de gran magnitud. Mientras Medicare estima pérdidas anuales de 17 000 millones de dólares por reingresos evitables, el costo total para el sistema sanitario estadounidense supera los 52 000 millones de dólares. Los modelos tradicionales, como el índice LACE, el HOSPITAL score o la regresión logística, alcanzan áreas bajo la curva (AUC) modestas de entre 0,60 y 0,72, niveles que resultan insuficientes para la toma de decisiones clínicas efectivas.
Este trabajo propone y evalúa una arquitectura híbrida CNN-LSTM que integra datos clínicos estáticos (demografía, comorbilidades de Charlson y diagnósticos ICD) con series temporales de signos vitales de las últimas 48 horas previas al alta, extraídas de MIMIC-IV. Se emplean convoluciones 1D para capturar patrones locales de deterioro y capas LSTM bidireccionales para modelar dependencias temporales largas, mediante una fusión intermedia de embeddings tabulares. Utilizando focal loss para gestionar el desbalance de datos (15 % de clase positiva) y validación estratificada, el modelo alcanza un AUC-ROC de 0,812 y un AUC-PR de 0,451, superando a XGBoost (0,742) y a una LSTM simple (0,776). Finalmente, se discute la interpretabilidad mediante SHAP, las implicaciones éticas y la aplicabilidad de esta tecnología al contexto latinoamericano.
Descargas
Citas
Aracena, C., Villena, F., Arias, F., & Dunstan, J. (2022). Aplicaciones de aprendizaje automático en salud. Revista Médica Clínica Las Condes, 33(6), 568–575. https://doi.org/10.1016/j.rmclc.2022.11.002
Ashfaq, A., Sant'Anna, A., Lingman, M., & Nowaczyk, S. (2019). Readmission prediction using deep learning on electronic health records. Journal of Biomedical Informatics, 97, 103256. https://doi.org/10.1016/j.jbi.2019.103256
Baker, S., Xiang, W., & Atkinson, I. (2020). Continuous and automatic mortality risk prediction using vital signs in the intensive care unit: A hybrid neural network approach. Scientific Reports, 10(1), 21282. https://doi.org/10.1038/s41598-020-78184-7
Barbieri, S., Kemp, J., Perez-Concha, O., Kotwal, S., Gallagher, M., Ritchie, A., & Jorm, L. (2020). Benchmarking deep learning architectures for predicting readmission to the ICU and describing patients-at-risk. Scientific Reports, 10(1), 1111. https://doi.org/10.1038/s41598-020-58053-z
Billings, J., Blunt, I., Steventon, A., Georghiou, T., Lewis, G., & Bardsley, M. (2012). Development of a predictive model to identify inpatients at risk of re-admission within 30 days of discharge (PARR-30). BMJ Open, 2(4), e001667. https://doi.org/10.1136/bmjopen-2012-001667
Cardona-Morrell, M., Prgomet, M., Turner, R. M., Nicholson, M., & Hillman, K. (2016). Frecuencia de reingresos hospitalarios y factores asociados en afiliados a una administradora de servicios de salud en Colombia. Cadernos de Saúde Pública, 32(7), e00146014. https://doi.org/10.1590/0102-311X00146014
Castro, M. S. M., Carvalho, M. S., & Travassos, C. (2005). Factors associated with readmission to a general hospital in Brazil. Cadernos de Saúde Pública, 21(4), 1186–1200. https://doi.org/10.1590/S0102-311X2005000400021
Castro G., P., Vukasovic R., J. L., Garcés S., E., Sepúlveda M., L., Ferrada K., M., & Alvarado O., S. (2004). Insuficiencia cardíaca en hospitales chilenos: resultados del Registro Nacional de Insuficiencia Cardíaca, Grupo ICARO. Revista Médica de Chile, 132(6), 655–662. https://doi.org/10.4067/S0034-98872004000600001
Char, D. S., Shah, N. H., & Magnus, D. (2018). Implementing machine learning in health care — Addressing ethical challenges. New England Journal of Medicine, 378(11), 981–983. https://doi.org/10.1056/NEJMp1714229
Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357. https://doi.org/10.1613/jair.953
Che, Z., Purushotham, S., Cho, K., Sontag, D., & Liu, Y. (2018). Recurrent neural networks for multivariate time series with missing values. Scientific Reports, 8(1), 6085. https://doi.org/10.1038/s41598-018-24271-9
Chen, I. Y., Pierson, E., Rose, S., Joshi, S., Ferryman, K., & Ghassemi, M. (2021). Ethical machine learning in healthcare. Annual Review of Biomedical Data Science, 4, 123–144. https://doi.org/10.1146/annurev-biodatasci-092820-114757
Choi, E., Bahadori, M. T., Kulas, J. A., Schuetz, A., Stewart, W. F., & Sun, J. (2016). RETAIN: An interpretable predictive model for healthcare using reverse time attention mechanism. Advances in Neural Information Processing Systems, 29, 3504–3512.
Chung, C. C., Chiu, I. M., Cheng, C. Y., Lin, C. H. R., Kung, C. T., Lin, Y. C., & Wu, C. C. (2022). Machine learning models for predicting in-hospital mortality in patient with sepsis: Analysis of vital sign dynamics. Frontiers in Medicine, 9, 964667. https://doi.org/10.3389/fmed.2022.964667
Collins, G. S., Moons, K. G. M., Dhiman, P., Riley, R. D., Beam, A. L., Van Calster, B., Ghassemi, M., Liu, X., Reitsma, J. B., van Smeden, M., Boulesteix, A.-L., Camaradou, J. C., Celi, L. A., Denaxas, S., Denniston, A. K., Glocker, B., Golub, R. M., Harvey, H., Heinze, G., … Logullo, P. (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
D'Agostino, M., Marti, M., Otero, P., De Cosio, F. G., Savignano, C., Nelson, J., García Saisó, S., Luna, D., Mayer-Lima, W. A., & Curioso, W. H. (2023). Digital transformation as a strategy to strengthen essential public health functions in the Americas. Revista Panamericana de Salud Pública, 47, e166. https://doi.org/10.26633/RPSP.2023.166
Davis, J., & Goadrich, M. (2006). The relationship between precision-recall and ROC curves. Proceedings of the 23rd International Conference on Machine Learning, 233–240. https://doi.org/10.1145/1143844.1143874
Delgado-Osorio, N., Vidal-Medina, U., López, D. M., & Blobel, B. (2023). Aprendizaje automático aplicado a la predicción de diabetes mellitus, utilizando información socioeconómica y ambiental de usuarios del sistema de salud. Revista Facultad Nacional de Salud Pública, 41(2), e351168. https://doi.org/10.17533/udea.rfnsp.e351168
Donzé, J., Aujesky, D., Williams, D., & Schnipper, J. L. (2013). Potentially avoidable 30-day hospital readmissions in medical patients: Derivation and validation of a prediction model. JAMA Internal Medicine, 173(8), 632–638. https://doi.org/10.1001/jamainternmed.2013.3023
Donzé, J. D., Williams, M. V., Robinson, E. J., Zimlichman, E., Aujesky, D., Vasilevskis, E. E., Kripalani, S., Metlay, J. P., Wallington, T., Fletcher, G. S., Auerbach, A. D., & Schnipper, J. L. (2016). International validity of the HOSPITAL score to predict 30-day potentially avoidable hospital readmissions. JAMA Internal Medicine, 176(4), 496–502. https://doi.org/10.1001/jamainternmed.2015.8462
Fierro Mella, C. C. (2019). Diseño y desarrollo de modelo predictivo de re-hospitalizaciones no planificadas usando Deep Learning [Tesis de pregrado, Universidad de Chile]. Repositorio Académico Universidad de Chile. https://repositorio.uchile.cl/handle/2250/174955
Futoma, J., Simons, M., Panch, T., Doshi-Velez, F., & Celi, L. A. (2020). The myth of generalisability in clinical research and machine learning in health care. The Lancet Digital Health, 2(9), e489–e492. https://doi.org/10.1016/S2589-7500(20)30186-2
Gianfrancesco, M. A., Tamang, S., Yazdany, J., & Schmajuk, G. (2018). Potential biases in machine learning algorithms using electronic health records. JAMA Internal Medicine, 178(11), 1544–1547. https://doi.org/10.1001/jamainternmed.2018.3763
Gutiérrez-Portela, F., Almenares-Mendoza, F., Calderón-Benavides, L., & Romero-Riaño, E. (2024). Predicción de riesgo cardiovascular mediante aprendizaje automático: un análisis comparativo entre técnicas. Ingeniería y Competitividad, 26(1). https://doi.org/10.25100/iyc.v26i1.13229
Harutyunyan, H., Khachatrian, H., Kale, D. C., Ver Steeg, G., & Galstyan, A. (2019). Multitask learning and benchmarking with clinical time series data. Scientific Data, 6(1), 96. https://doi.org/10.1038/s41597-019-0103-9
Hayat, N., Geras, K. J., & Shamout, F. E. (2022). MedFuse: Multi-modal fusion with clinical time-series data and chest X-ray images. Proceedings of Machine Learning Research, 182, 479–503.
Huang, K., Altosaar, J., & Ranganath, R. (2020). ClinicalBERT: Modeling clinical notes and predicting hospital readmission. Proceedings of the ACM Conference on Health, Inference, and Learning Workshop. https://doi.org/10.48550/arXiv.1904.05342
Huang, S.-C., Pareek, A., Seyyedi, S., Banerjee, I., & Lungren, M. P. (2020). Fusion of medical imaging and electronic health records using deep learning: A systematic review and implementation guidelines. npj Digital Medicine, 3(1), 136. https://doi.org/10.1038/s41746-020-00341-z
Hyland, S. L., Faltys, M., Hüser, M., Lyu, X., Gumbsch, T., Esteban, C., Bock, C., Horn, M., Moor, M., Rieck, B., Zimmermann, M., Bodenham, D., Borgwardt, K., Rätsch, G., & Merz, T. M. (2020). Early prediction of circulatory failure in the intensive care unit using machine learning. Nature Medicine, 26(3), 364–373. https://doi.org/10.1038/s41591-020-0789-4
Jencks, S. F., Williams, M. V., & Coleman, E. A. (2009). Rehospitalizations among patients in the Medicare fee-for-service program. New England Journal of Medicine, 360(14), 1418–1428. https://doi.org/10.1056/NEJMsa0803563
Johnson, A. E. W., Bulgarelli, L., Shen, L., Gayles, A., Shammout, A., Horng, S., Pollard, T. J., Hao, S., Moody, B., Gow, B., Lehman, L. H., Celi, L. A., & Mark, R. G. (2023). MIMIC-IV, a freely accessible electronic health record dataset. Scientific Data, 10(1), 1. https://doi.org/10.1038/s41597-022-01899-x
Kansagara, D., Englander, H., Salanitro, A., Kagen, D., Theobald, C., Freeman, M., & Kripalani, S. (2011). Risk prediction models for hospital readmission: A systematic review. JAMA, 306(15), 1688–1698. https://doi.org/10.1001/jama.2011.1515
Khan, M. S., Sreenivasan, J., Lateef, N., Abougergi, M. S., Greene, S. J., Ahmad, T., Anker, S. D., Fonarow, G. C., & Butler, J. (2021). Trends in 30- and 90-day readmission rates for heart failure. Circulation: Heart Failure, 14(4), e008335. https://doi.org/10.1161/CIRCHEARTFAILURE.121.008335
Khan, M. S., Samman Tahhan, A., Vaduganathan, M., Greene, S. J., Alrohaibani, A., Anker, S. D., Fonarow, G. C., & Butler, J. (2023). Global comparison of readmission rates for patients with heart failure. Journal of the American College of Cardiology, 82(5), 430–444. https://doi.org/10.1016/j.jacc.2023.05.040
Khera, R., Wang, Y., Bernheim, S. M., Lin, Z., & Krumholz, H. M. (2020). Post-discharge acute care and outcomes following readmission reduction initiatives. BMJ, 368, l6831. https://doi.org/10.1136/bmj.l6831
Kingma, D. P., & Ba, J. (2015). Adam: A method for stochastic optimization. Proceedings of the 3rd International Conference on Learning Representations. https://doi.org/10.48550/arXiv.1412.6980
Li, Y., Rao, S., Solares, J. R. A., Hassaine, A., Ramakrishnan, R., Canoy, D., Zhu, Y., Rahimi, K., & Salimi-Khorshidi, G. (2020). BEHRT: Transformer for electronic health records. Scientific Reports, 10(1), 7155. https://doi.org/10.1038/s41598-020-62922-y
Lin, T.-Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017). Focal loss for dense object detection. Proceedings of the IEEE International Conference on Computer Vision, 2980–2988. https://doi.org/10.1109/ICCV.2017.324
Lin, Y.-W., Zhou, Y., Faghri, F., Shaw, M. J., & Campbell, R. H. (2019). Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory. PLOS ONE, 14(7), e0218942. https://doi.org/10.1371/journal.pone.0218942
Lipton, Z. C., Kale, D. C., Elkan, C., & Wetzel, R. (2016). Learning to diagnose with LSTM recurrent neural networks. Proceedings of the 4th International Conference on Learning Representations. https://doi.org/10.48550/arXiv.1511.03677
López, D. M., & Osorio, J. S. (2025). Inteligencia artificial y salud digital en Colombia: panorama de los avances más recientes y retos futuros. Biomédica, 45(Sp. 3), 5–9. https://doi.org/10.7705/biomedica.8248
Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765–4774.
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S.-I. (2020). From local explanations to global understanding with explainable AI for trees. Nature Machine Intelligence, 2(1), 56–67. https://doi.org/10.1038/s42256-019-0138-9
Mahmoudi, E., Kamdar, N., Kim, N., Gonzales, G., Singh, K., & Waljee, A. K. (2020). Use of electronic medical records in development and validation of risk prediction models of hospital readmission: Systematic review. BMJ, 369, m958. https://doi.org/10.1136/bmj.m958
Miotto, R., Li, L., Kidd, B. A., & Dudley, J. T. (2016). Deep Patient: An unsupervised representation to predict the future of patients from the electronic health records. Scientific Reports, 6, 26094. https://doi.org/10.1038/srep26094
Miotto, R., Wang, F., Wang, S., Jiang, X., & Dudley, J. T. (2018). Deep learning for healthcare: Review, opportunities and challenges. Briefings in Bioinformatics, 19(6), 1236–1246. https://doi.org/10.1093/bib/bbx044
Moor, M., Banerjee, O., Abad, Z. S. H., Krumholz, H. M., Leskovec, J., Topol, E. J., & Rajpurkar, P. (2023). Foundation models for generalist medical artificial intelligence. Nature, 616(7956), 259–265. https://doi.org/10.1038/s41586-023-05881-4
Myers, L. C., Faridi, M. K., Currier, P., Camargo, C. A., & Hasegawa, K. (2020). The Hospital Readmissions Reduction Program and readmissions for chronic obstructive pulmonary disease, 2006–2015. Annals of the American Thoracic Society, 17(4), 450–456. https://doi.org/10.1513/AnnalsATS.201909-672OC
Nazzal N., C., Campos T., P., Corbalán H., R., Lanas Z., F., Bartolucci J., J., Sanhueza C., P., Cavada Ch., G., & Prieto D., J. C. (2017). Incidencia y letalidad intrahospitalaria por insuficiencia cardiaca en Chile: ¿Existen diferencias por sexo? Revista Médica de Chile, 145(6), 703–709. https://doi.org/10.4067/s0034-98872017000600703
Norgeot, B., Quer, G., Beaulieu-Jones, B. K., Torkamani, A., Dias, R., Gianfrancesco, M., Arnaout, R., Kohane, I. S., Saria, S., Topol, E., Obermeyer, Z., Yu, B., & Butte, A. J. (2020). Minimum information about clinical artificial intelligence modeling: The MI-CLAIM checklist. Nature Medicine, 26(9), 1320–1324. https://doi.org/10.1038/s41591-020-1041-y
Obermeyer, Z., Powers, B., Vogeli, C., & Mullainathan, S. (2019). Dissecting racial bias in an algorithm used to manage the health of populations. Science, 366(6464), 447–453. https://doi.org/10.1126/science.aax2342
Organización Panamericana de la Salud. (2023). Caja de herramientas para la transformación digital del sector salud (Versión 2.0). OPS/OMS.
Pollard, T. J., Johnson, A. E. W., Raffa, J. D., Celi, L. A., Mark, R. G., & Badawi, O. (2018). The eICU Collaborative Research Database, a freely available multi-center database for critical care research. Scientific Data, 5(1), 180178. https://doi.org/10.1038/sdata.2018.178
Purushotham, S., Meng, C., Che, Z., & Liu, Y. (2018). Benchmarking deep learning models on large healthcare datasets. Journal of Biomedical Informatics, 83, 112–134. https://doi.org/10.1016/j.jbi.2018.04.007
Rafiei, A., Rezaee, A., Hajati, F., Gheisari, S., & Golzan, M. (2021). SSP: Early prediction of sepsis using fully connected LSTM-CNN model. Computers in Biology and Medicine, 128, 104110. https://doi.org/10.1016/j.compbiomed.2020.104110
Rajkomar, A., Hardt, M., Howell, M. D., Corrado, G., & Chin, M. H. (2018). Ensuring fairness in machine learning to advance health equity. Annals of Internal Medicine, 169(12), 866–872. https://doi.org/10.7326/M18-1990
Rajkomar, A., Oren, E., Chen, K., Dai, A. M., Hajaj, N., Hardt, M., Liu, P. J., Liu, X., Marcus, J., Sun, M., Sundberg, P., Yee, H., Zhang, K., Zhang, Y., Flores, G., Duggan, G. E., Irvine, J., Le, Q., Litsch, K., … Dean, J. (2018). Scalable an
Al Alawi, A., et al. (2025). Development and evaluation of a machine learning model for predicting 30-day readmission in general internal medicine. Computers, 14(5), 177. https://doi.org/10.3390/computers14050177
Alemán Flores, M., & Ynojosa Parra, L. (2024). Desafíos frente al uso de las TICs en pandemia para el programa de enfermería. Zenodo. https://doi.org/10.5281/zenodo.10465777
Cabrera-Mudarra, J. (2025). Inteligencia artificial y calidad clínica: Herramientas para optimizar el desempeño hospitalario. Noesis, 7(14). https://doi.org/10.35381/noesisin.v7i14.349
Hedayet, N., & Haseen, M. (2024). Artificial intelligence in public health: A review article. Bangladesh Journal of Bioethics, 15(2). https://doi.org/10.62865/bjbio.v15i2.108
Lee, C., et al. (2023). Multimodal deep learning of fundus abnormalities and traditional risk factors for cardiovascular risk prediction. NPJ Digital Medicine, 6, 48. https://doi.org/10.1038/s41746-023-00748-4
Liu, Y., et al. (2023). Learning multi-modal brain tumor segmentation from MRI images. Computers in Biology and Medicine, 158, 106927. https://doi.org/10.1016/j.compbiomed.2023.106927
López-Canay, A., et al. (2025). Predicting COPD readmission: An intelligent clinical decision support system. Diagnostics, 15(3), 318. https://doi.org/10.3390/diagnostics15030318
Mahootiha, S., et al. (2024). Multimodal deep learning for renal carcinoma prognosis. Computer Methods and Programs in Biomedicine, 240, 107978. https://doi.org/10.1016/j.cmpb.2023.107978
Mohsen, H., et al. (2022). Artificial intelligence-based methods for fusion of EHR and imaging data. Scientific Reports, 12, 22514. https://doi.org/10.1038/s41598-022-22514-4
Ruiz, J., & Velásquez, M. (2023). Inteligencia artificial al servicio de la salud del futuro. Revista Médica Clínica Las Condes, 34(1). https://doi.org/10.1016/j.rmclc.2022.12.001
Descargas
Publicado
Cómo citar
Número
Sección
Licencia
Derechos de autor 2026 Treisy Alexandra Castillo Camacho, María Ines Cevallos Mazon, Flor Gabriela Soliz Poveda, Alysson Valeria Urrea Camacho, Andrea Coraima Espin Lasso, Amanda Cristina Carlosama Calderon
Esta obra está bajo una licencia internacional Creative Commons Atribución 4.0.
Los autores conservan los derechos morales y patrimoniales de sus obras. Puesto que Salud ConCiencia es una publicación de acceso abierto, los lectores pueden reproducir total o parcialmente su contenido siempre y cuando proporcionen adecuadamente el crédito a los autores correspondientes y a la revista misma.
Nuestra revista se rige por las políticas internacionales SHERPA/RoMEO: Revista verde: Permiten el autoarchivo tanto del pre-print (borrador de un trabajo) como del post-print (la versión corregida y revisada por pares) y hasta de la versión final (maquetada tal como saldrá publicada en la revista).
Véase también "Derechos de autor y licencias".
