[Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years
Background and objective: The objective is to develop a model that predicts vital status six months after fracture as accurately as possible. For this purpose we will use five different data sources obtained through the National Hip Fracture Registry, the Health Management Unit and the Economic Mana...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-01-01
|
| Series: | Revista Española de Cirugía Ortopédica y Traumatología |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S1888441524001838 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841553829589417984 |
|---|---|
| author | I. Calvo Lorenzo I. Uriarte Llano M.R. Mateo Citores Y. Rojo Maza U. Agirregoitia Enzunza |
| author_facet | I. Calvo Lorenzo I. Uriarte Llano M.R. Mateo Citores Y. Rojo Maza U. Agirregoitia Enzunza |
| author_sort | I. Calvo Lorenzo |
| collection | DOAJ |
| description | Background and objective: The objective is to develop a model that predicts vital status six months after fracture as accurately as possible. For this purpose we will use five different data sources obtained through the National Hip Fracture Registry, the Health Management Unit and the Economic Management Department. Material and methods: The study population is a cohort of patients over 74 years of age who suffered a hip fracture between May 2020 and December 2022. A warehouse is created from five different data sources with the necessary variables. An analysis of missing values and outliers as well as unbalanced classes of the target variable (“vital status”) is performed. Fourteen different algorithmic models are trained with the training. The model with the best performance is selected and a fine tuning is performed. Finally, the performance of the selected model is analysed with test data. Results: A data warehouse is created with 502 patients and 144 variables. The best performing model is Linear Regression. Sixteen of the 24 cases of deceased patients are classified as live, and 14 live patients are classified as deceased. A sensitivity of 31%, an accuracy of 34% and an area under the curve of 0.65 is achieved. Conclusions: We have not been able to generate a model for the prediction of six-month survival in the current cohort. However, we believe that the method used for the generation of algorithms based on machine learning can serve as a reference for future works. Resumen: Objetivo: Desarrollar un modelo que prediga con la mayor exactitud posible el estado vital a los 6 meses tras fractura de cadera, utilizando para ello cinco fuentes de datos obtenidas a través del Registro Nacional de Fracturas de Cadera, la Unidad de Gestión Sanitaria y la Dirección Económica. Material y metodología: La población de estudio es una cohorte de pacientes que sufrieron fractura de cadera entre mayo de 2020 y diciembre de 2022. A partir de cinco fuentes diferentes de datos se crea un almacén con las variables necesarias. Se realiza un análisis de valores perdidos y atípicos, así como de desbalanceo de las clases de la variable objetivo («estado vital»). Se entrenan 14 diferentes modelos algorítmicos con los datos de entrenamiento. Se selecciona el modelo que mejor rendimiento obtenga y se realiza una puesta a punto fina. Finalmente se analiza el rendimiento del modelo con datos de test. Resultados: Se crea un almacén de datos con 502 pacientes y 144 variables. El modelo con mejor rendimiento es la regresión lineal. Dieciséis de los 24 casos de pacientes fallecidos son clasificados como vivos, y 14 pacientes vivos son clasificados como fallecidos. Se consigue una sensibilidad del 31%, una precisión del 34% y un área bajo la curva de 0,65. Conclusiones: No se ha conseguido generar un modelo de predicción de muerte a los 6 meses con nuestra cohorte. Sin embargo, creemos que el método utilizado para generar algoritmos basados en aprendizaje automático puede servir de referencia para futuros trabajos. |
| format | Article |
| id | doaj-art-814d3141400542feaa50f3a15e42f5cc |
| institution | Kabale University |
| issn | 1888-4415 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Revista Española de Cirugía Ortopédica y Traumatología |
| spelling | doaj-art-814d3141400542feaa50f3a15e42f5cc2025-01-09T06:13:42ZengElsevierRevista Española de Cirugía Ortopédica y Traumatología1888-44152025-01-01691T47T54[Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 yearsI. Calvo Lorenzo0I. Uriarte Llano1M.R. Mateo Citores2Y. Rojo Maza3U. Agirregoitia Enzunza4Corresponding author.; Servicio de Cirugía Ortopédica y Traumatología, Hospital Universitario Galdakao-Usansolo, Galdakao, Bizkaia, SpainServicio de Cirugía Ortopédica y Traumatología, Hospital Universitario Galdakao-Usansolo, Galdakao, Bizkaia, SpainServicio de Cirugía Ortopédica y Traumatología, Hospital Universitario Galdakao-Usansolo, Galdakao, Bizkaia, SpainServicio de Cirugía Ortopédica y Traumatología, Hospital Universitario Galdakao-Usansolo, Galdakao, Bizkaia, SpainServicio de Cirugía Ortopédica y Traumatología, Hospital Universitario Galdakao-Usansolo, Galdakao, Bizkaia, SpainBackground and objective: The objective is to develop a model that predicts vital status six months after fracture as accurately as possible. For this purpose we will use five different data sources obtained through the National Hip Fracture Registry, the Health Management Unit and the Economic Management Department. Material and methods: The study population is a cohort of patients over 74 years of age who suffered a hip fracture between May 2020 and December 2022. A warehouse is created from five different data sources with the necessary variables. An analysis of missing values and outliers as well as unbalanced classes of the target variable (“vital status”) is performed. Fourteen different algorithmic models are trained with the training. The model with the best performance is selected and a fine tuning is performed. Finally, the performance of the selected model is analysed with test data. Results: A data warehouse is created with 502 patients and 144 variables. The best performing model is Linear Regression. Sixteen of the 24 cases of deceased patients are classified as live, and 14 live patients are classified as deceased. A sensitivity of 31%, an accuracy of 34% and an area under the curve of 0.65 is achieved. Conclusions: We have not been able to generate a model for the prediction of six-month survival in the current cohort. However, we believe that the method used for the generation of algorithms based on machine learning can serve as a reference for future works. Resumen: Objetivo: Desarrollar un modelo que prediga con la mayor exactitud posible el estado vital a los 6 meses tras fractura de cadera, utilizando para ello cinco fuentes de datos obtenidas a través del Registro Nacional de Fracturas de Cadera, la Unidad de Gestión Sanitaria y la Dirección Económica. Material y metodología: La población de estudio es una cohorte de pacientes que sufrieron fractura de cadera entre mayo de 2020 y diciembre de 2022. A partir de cinco fuentes diferentes de datos se crea un almacén con las variables necesarias. Se realiza un análisis de valores perdidos y atípicos, así como de desbalanceo de las clases de la variable objetivo («estado vital»). Se entrenan 14 diferentes modelos algorítmicos con los datos de entrenamiento. Se selecciona el modelo que mejor rendimiento obtenga y se realiza una puesta a punto fina. Finalmente se analiza el rendimiento del modelo con datos de test. Resultados: Se crea un almacén de datos con 502 pacientes y 144 variables. El modelo con mejor rendimiento es la regresión lineal. Dieciséis de los 24 casos de pacientes fallecidos son clasificados como vivos, y 14 pacientes vivos son clasificados como fallecidos. Se consigue una sensibilidad del 31%, una precisión del 34% y un área bajo la curva de 0,65. Conclusiones: No se ha conseguido generar un modelo de predicción de muerte a los 6 meses con nuestra cohorte. Sin embargo, creemos que el método utilizado para generar algoritmos basados en aprendizaje automático puede servir de referencia para futuros trabajos.http://www.sciencedirect.com/science/article/pii/S1888441524001838Fractura de caderaAprendizaje automáticoBases de datosNoSQL |
| spellingShingle | I. Calvo Lorenzo I. Uriarte Llano M.R. Mateo Citores Y. Rojo Maza U. Agirregoitia Enzunza [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years Revista Española de Cirugía Ortopédica y Traumatología Fractura de cadera Aprendizaje automático Bases de datos NoSQL |
| title | [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| title_full | [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| title_fullStr | [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| title_full_unstemmed | [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| title_short | [Translated article] Analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| title_sort | translated article analysis of machine learning algorithmic models for the prediction of vital status at six months after hip fracture in patients older than 74 years |
| topic | Fractura de cadera Aprendizaje automático Bases de datos NoSQL |
| url | http://www.sciencedirect.com/science/article/pii/S1888441524001838 |
| work_keys_str_mv | AT icalvolorenzo translatedarticleanalysisofmachinelearningalgorithmicmodelsforthepredictionofvitalstatusatsixmonthsafterhipfractureinpatientsolderthan74years AT iuriartellano translatedarticleanalysisofmachinelearningalgorithmicmodelsforthepredictionofvitalstatusatsixmonthsafterhipfractureinpatientsolderthan74years AT mrmateocitores translatedarticleanalysisofmachinelearningalgorithmicmodelsforthepredictionofvitalstatusatsixmonthsafterhipfractureinpatientsolderthan74years AT yrojomaza translatedarticleanalysisofmachinelearningalgorithmicmodelsforthepredictionofvitalstatusatsixmonthsafterhipfractureinpatientsolderthan74years AT uagirregoitiaenzunza translatedarticleanalysisofmachinelearningalgorithmicmodelsforthepredictionofvitalstatusatsixmonthsafterhipfractureinpatientsolderthan74years |