Early Prediction of ICU Readmissions Using Machine Learning in a Colombian University Hospital / José Leonardo Mojica Peñaranda ; Camacho Cocollo, J.E ; Lerma Pazos, Joel Andrés  

Público ISBD Título : Early Prediction of ICU Readmissions Using Machine Learning in a Colombian University Hospital Tipo de documento : documento electrónico Autores : José Leonardo Mojica Peñaranda, Autor ; Camacho Cocollo, J.E, Autor ; Lerma Pazos, Joel Andrés, Autor Fecha de publicación : 2025 Títulos uniformes : Springer Idioma : Inglés (eng) Resumen : Hospital readmissions are common, unplanned, and potentially avoidable occurrences linked to elevated morbidity and mortality rates. In Colombia, the primary concerns regarding ICU readmissions involve the significant expenses associated with treating patients readmitted due to deteriorating circumstances or heightened illness severity, alongside inadequate discharge decisions made by certain physicians or experts, which decrease the efficacy of healthcare facilities. This study aims to develop a predictive model that provides an early warning for the readmission of adult patients to the intensive care unit. In order to accomplish this, a variety of database administration and data analysis technologies were employed, alongside with machine learning models referenced in the literature. These models were trained and validated using data provided by the Hospital Pablo Tobón Uribe (HPTU) in Medellín. As a result of this project, a predictive model was developed with an accuracy score of 0.74 and an AUC value of 0.74, providing valuable information to physicians and specialists when making ICU discharge decisions. This model generates alerts for patients at risk of readmission if discharged, improving decision-making in intensive care settings. Mención de responsabilidad : Joel Andrés Lerma Pazos, J. E. Camacho Cogollo, Jose Leonardo Mojica Peñaranda DOI (Digital Object Identifier) : 10.1007/978-3-031-96538-8_11 Derechos de uso : CC BY-NC-ND En línea : https://link.springer.com/chapter/10.1007/978-3-031-96538-8_11 Enlace permanente : https://hospitalpablotobon.cloudbiteca.com/pmb/opac_css/index.php?lvl=notice_dis Early Prediction of ICU Readmissions Using Machine Learning in a Colombian University Hospital [documento electrónico] / José Leonardo Mojica Peñaranda, Autor ; Camacho Cocollo, J.E, Autor ; Lerma Pazos, Joel Andrés, Autor . - 2025. Obra : Springer Idioma : Inglés (eng) Resumen : Hospital readmissions are common, unplanned, and potentially avoidable occurrences linked to elevated morbidity and mortality rates. In Colombia, the primary concerns regarding ICU readmissions involve the significant expenses associated with treating patients readmitted due to deteriorating circumstances or heightened illness severity, alongside inadequate discharge decisions made by certain physicians or experts, which decrease the efficacy of healthcare facilities. This study aims to develop a predictive model that provides an early warning for the readmission of adult patients to the intensive care unit. In order to accomplish this, a variety of database administration and data analysis technologies were employed, alongside with machine learning models referenced in the literature. These models were trained and validated using data provided by the Hospital Pablo Tobón Uribe (HPTU) in Medellín. As a result of this project, a predictive model was developed with an accuracy score of 0.74 and an AUC value of 0.74, providing valuable information to physicians and specialists when making ICU discharge decisions. This model generates alerts for patients at risk of readmission if discharged, improving decision-making in intensive care settings. Mención de responsabilidad : Joel Andrés Lerma Pazos, J. E. Camacho Cogollo, Jose Leonardo Mojica Peñaranda DOI (Digital Object Identifier) : 10.1007/978-3-031-96538-8_11 Derechos de uso : CC BY-NC-ND En línea : https://link.springer.com/chapter/10.1007/978-3-031-96538-8_11 Enlace permanente : https://hospitalpablotobon.cloudbiteca.com/pmb/opac_css/index.php?lvl=notice_dis

Reserva

Reservar este documento

Ejemplares(1)

Código de barras	Número de Ubicación	Tipo de medio	Ubicación	Sección	Estado
DD002379	AC-2025-062	Archivo digital	Producción Científica	Artículos científicos	Disponible

---

1 (1 - 1 / 1) Por página: 25 50 100 200