Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units
Objectives Early hemorrhage detection in intensive care units (ICUs) enables timely intervention and reduces the risk of irreversible outcomes. In this study, we aimed to develop a machine learning model to predict hemorrhage by learning the patterns of continuously changing, real-world clinical dat...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
The Korean Society of Medical Informatics
2022-10-01
|
| Series: | Healthcare Informatics Research |
| Subjects: | |
| Online Access: | http://www.e-hir.org/upload/pdf/hir-2022-28-4-364.pdf |
| _version_ | 1811317154700591104 |
|---|---|
| author | Sora Kang Chul Park Jinseok Lee Dukyong Yoon |
| author_facet | Sora Kang Chul Park Jinseok Lee Dukyong Yoon |
| author_sort | Sora Kang |
| collection | DOAJ |
| description | Objectives Early hemorrhage detection in intensive care units (ICUs) enables timely intervention and reduces the risk of irreversible outcomes. In this study, we aimed to develop a machine learning model to predict hemorrhage by learning the patterns of continuously changing, real-world clinical data. Methods We used the Medical Information Mart for Intensive Care databases (MIMIC-III and MIMIC-IV). A recurrent neural network was used to predict severe hemorrhage in the ICU. We developed three machine learning models with an increasing number of input features and levels of complexity: model 1 (11 features), model 2 (18 features), and model 3 (27 features). MIMIC-III was used for model training, and MIMIC-IV was split for internal validation. Using the model with the highest performance, external verification was performed using data from a subgroup extracted from the eICU Collaborative Research Database. Results We included 5,670 ICU admissions, with 3,150 in the training set and 2,520 in the internal test set. A positive correlation was found between model complexity and performance. As a measure of performance, three models developed with an increasing number of features showed area under the receiver operating characteristic (AUROC) curve values of 0.61–0.94 according to the range of input data. In the subgroup extracted from the eICU database for external validation, an AUROC value of 0.74 was observed. Conclusions Machine learning models that rely on real clinical data can be used to predict patients at high risk of bleeding in the ICU. |
| first_indexed | 2024-04-13T12:03:04Z |
| format | Article |
| id | doaj.art-bdeee946740c4e4eabfd2feb98079467 |
| institution | Directory Open Access Journal |
| issn | 2093-3681 2093-369X |
| language | English |
| last_indexed | 2024-04-13T12:03:04Z |
| publishDate | 2022-10-01 |
| publisher | The Korean Society of Medical Informatics |
| record_format | Article |
| series | Healthcare Informatics Research |
| spelling | doaj.art-bdeee946740c4e4eabfd2feb980794672022-12-22T02:47:44ZengThe Korean Society of Medical InformaticsHealthcare Informatics Research2093-36812093-369X2022-10-0128436437510.4258/hir.2022.28.4.3641139Machine Learning Model for the Prediction of Hemorrhage in Intensive Care UnitsSora Kang0Chul Park1Jinseok Lee2Dukyong Yoon3 Department of Biomedical Informatics, Ajou University School of Medicine, Suwon, Korea Division of Pulmonology, Department of Internal Medicine, Wonkwang University Hospital, Iksan, Korea Department of Biomedical Engineering, College of Electronics and Information, Kyung Hee University, Yongin, Korea Department of Biomedical Systems Informatics, Yonsei University College of Medicine, Yongin, KoreaObjectives Early hemorrhage detection in intensive care units (ICUs) enables timely intervention and reduces the risk of irreversible outcomes. In this study, we aimed to develop a machine learning model to predict hemorrhage by learning the patterns of continuously changing, real-world clinical data. Methods We used the Medical Information Mart for Intensive Care databases (MIMIC-III and MIMIC-IV). A recurrent neural network was used to predict severe hemorrhage in the ICU. We developed three machine learning models with an increasing number of input features and levels of complexity: model 1 (11 features), model 2 (18 features), and model 3 (27 features). MIMIC-III was used for model training, and MIMIC-IV was split for internal validation. Using the model with the highest performance, external verification was performed using data from a subgroup extracted from the eICU Collaborative Research Database. Results We included 5,670 ICU admissions, with 3,150 in the training set and 2,520 in the internal test set. A positive correlation was found between model complexity and performance. As a measure of performance, three models developed with an increasing number of features showed area under the receiver operating characteristic (AUROC) curve values of 0.61–0.94 according to the range of input data. In the subgroup extracted from the eICU database for external validation, an AUROC value of 0.74 was observed. Conclusions Machine learning models that rely on real clinical data can be used to predict patients at high risk of bleeding in the ICU.http://www.e-hir.org/upload/pdf/hir-2022-28-4-364.pdfhemorrhageprognosisintensive care unitsmonitoringphysiologicalblood transfusion |
| spellingShingle | Sora Kang Chul Park Jinseok Lee Dukyong Yoon Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units Healthcare Informatics Research hemorrhage prognosis intensive care units monitoring physiological blood transfusion |
| title | Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units |
| title_full | Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units |
| title_fullStr | Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units |
| title_full_unstemmed | Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units |
| title_short | Machine Learning Model for the Prediction of Hemorrhage in Intensive Care Units |
| title_sort | machine learning model for the prediction of hemorrhage in intensive care units |
| topic | hemorrhage prognosis intensive care units monitoring physiological blood transfusion |
| url | http://www.e-hir.org/upload/pdf/hir-2022-28-4-364.pdf |
| work_keys_str_mv | AT sorakang machinelearningmodelforthepredictionofhemorrhageinintensivecareunits AT chulpark machinelearningmodelforthepredictionofhemorrhageinintensivecareunits AT jinseoklee machinelearningmodelforthepredictionofhemorrhageinintensivecareunits AT dukyongyoon machinelearningmodelforthepredictionofhemorrhageinintensivecareunits |