Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning

Abstract Trauma-induced coagulopathy (TIC) has a high incidence in patients with severe trauma. Patients who develop TIC usually have a poor prognosis, characterised by increased organ dysfunction, susceptibility to sepsis, and high mortality. Nonetheless, there are still few studies specifically fo...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaohui Du, Wei Wang, Bo Xu, Jiang Zheng, Victor W. Xia, Yi Guo, Shuai Feng, Qingxiang Mao, Hong Fu
Format: Article
Language:English
Published: Nature Portfolio 2025-07-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-13283-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849766307718758400
author Xiaohui Du
Wei Wang
Bo Xu
Jiang Zheng
Victor W. Xia
Yi Guo
Shuai Feng
Qingxiang Mao
Hong Fu
author_facet Xiaohui Du
Wei Wang
Bo Xu
Jiang Zheng
Victor W. Xia
Yi Guo
Shuai Feng
Qingxiang Mao
Hong Fu
author_sort Xiaohui Du
collection DOAJ
description Abstract Trauma-induced coagulopathy (TIC) has a high incidence in patients with severe trauma. Patients who develop TIC usually have a poor prognosis, characterised by increased organ dysfunction, susceptibility to sepsis, and high mortality. Nonetheless, there are still few studies specifically focusing on postoperative TIC in severely traumatic patients. Therefore, the aim of this study was to construct a machine learning model for early identification of people at high risk of postoperative TIC. This retrospective analysis included data of severe trauma patients undergoing surgical treatment from January 2013 to February 2023 across four hospitals in China. Data of one hospital (n = 1204) was used for the development dataset, while other three hospitals contributed to the external validation dataset (n = 863). The study employed various machine learning algorithms, including random forests, logistic regression, gradient boosting decision trees, support vector machines, backpropagation artificial neural networks, extreme gradient boosting, and naïve Bayes. Model performance was estimated on the basis of accuracy, sensitivity, specificity, and area under the curve. In the internal cross-validation dataset, Shapley’s additive interpretation was applied to the model with the largest area under the receiver operating characteristic curve. TIC occurred in 25.4% (306/1204) and 2.9% (25/863) of patients in the developing and external validation set, respectively. Among the models evaluated, the Random Forest model demonstrated the highest performance, achieving an area under the curve of 0.82 for the test cohort and 0.73 for the external validation cohort. The findings suggest that machine learning models can effectively identify severely traumatized patients at a higher risk of postoperative trauma-induced coagulopathy. Utilizing machine learning may enhance clinical decision-making and improve management strategies for postoperative coagulation issues.
format Article
id doaj-art-05d9d5a09e854b52a885c915e0f4de6e
institution DOAJ
issn 2045-2322
language English
publishDate 2025-07-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-05d9d5a09e854b52a885c915e0f4de6e2025-08-20T03:04:37ZengNature PortfolioScientific Reports2045-23222025-07-0115111710.1038/s41598-025-13283-xPredicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learningXiaohui Du0Wei Wang1Bo Xu2Jiang Zheng3Victor W. Xia4Yi Guo5Shuai Feng6Qingxiang Mao7Hong Fu8Department of Anesthesiology, Chongqing Emergency Medical Center, School of Medicine, Chongqing University Central Hospital, Chongqing UniversityDepartment of Anesthesiology, The PLA Rocket Force Characteristic Medical CenterDepartment of Anesthesiology, General Hospital of Southern Theater Command of PLADepartment of Anesthesiology, Chongqing Emergency Medical Center, School of Medicine, Chongqing University Central Hospital, Chongqing UniversityDepartment of Anesthesiology, David Geffen School of Medicine, University of California, Los AngelesDepartment of Anesthesiology, The PLA Rocket Force Characteristic Medical CenterDepartment of Anesthesiology, The PLA Rocket Force Characteristic Medical CenterDepartment of Anesthesiology, Daping Hospital, Army Medical UniversityDepartment of Anesthesiology, Chongqing Emergency Medical Center, School of Medicine, Chongqing University Central Hospital, Chongqing UniversityAbstract Trauma-induced coagulopathy (TIC) has a high incidence in patients with severe trauma. Patients who develop TIC usually have a poor prognosis, characterised by increased organ dysfunction, susceptibility to sepsis, and high mortality. Nonetheless, there are still few studies specifically focusing on postoperative TIC in severely traumatic patients. Therefore, the aim of this study was to construct a machine learning model for early identification of people at high risk of postoperative TIC. This retrospective analysis included data of severe trauma patients undergoing surgical treatment from January 2013 to February 2023 across four hospitals in China. Data of one hospital (n = 1204) was used for the development dataset, while other three hospitals contributed to the external validation dataset (n = 863). The study employed various machine learning algorithms, including random forests, logistic regression, gradient boosting decision trees, support vector machines, backpropagation artificial neural networks, extreme gradient boosting, and naïve Bayes. Model performance was estimated on the basis of accuracy, sensitivity, specificity, and area under the curve. In the internal cross-validation dataset, Shapley’s additive interpretation was applied to the model with the largest area under the receiver operating characteristic curve. TIC occurred in 25.4% (306/1204) and 2.9% (25/863) of patients in the developing and external validation set, respectively. Among the models evaluated, the Random Forest model demonstrated the highest performance, achieving an area under the curve of 0.82 for the test cohort and 0.73 for the external validation cohort. The findings suggest that machine learning models can effectively identify severely traumatized patients at a higher risk of postoperative trauma-induced coagulopathy. Utilizing machine learning may enhance clinical decision-making and improve management strategies for postoperative coagulation issues.https://doi.org/10.1038/s41598-025-13283-xSeverely traumaTrauma-induce coagulopathyMachine learning
spellingShingle Xiaohui Du
Wei Wang
Bo Xu
Jiang Zheng
Victor W. Xia
Yi Guo
Shuai Feng
Qingxiang Mao
Hong Fu
Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
Scientific Reports
Severely trauma
Trauma-induce coagulopathy
Machine learning
title Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
title_full Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
title_fullStr Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
title_full_unstemmed Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
title_short Predicting postoperative trauma-induced coagulopathy in patients with severe injuries by machine learning
title_sort predicting postoperative trauma induced coagulopathy in patients with severe injuries by machine learning
topic Severely trauma
Trauma-induce coagulopathy
Machine learning
url https://doi.org/10.1038/s41598-025-13283-x
work_keys_str_mv AT xiaohuidu predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT weiwang predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT boxu predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT jiangzheng predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT victorwxia predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT yiguo predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT shuaifeng predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT qingxiangmao predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning
AT hongfu predictingpostoperativetraumainducedcoagulopathyinpatientswithsevereinjuriesbymachinelearning