Prediction of Postoperative Pulmonary Edema Risk Using Machine Learning
-
Published:2023-02-23
Issue:5
Volume:12
Page:1804
-
ISSN:2077-0383
-
Container-title:Journal of Clinical Medicine
-
language:en
-
Short-container-title:JCM
Author:
Kim Jong12ORCID, Kim Youngmi2, Yoo Kookhyun1, Kim Minguan1, Kang Seong3, Kwon Young-Suk12ORCID, Lee Jae12
Affiliation:
1. Department of Anesthesiology and Pain Medicine, Chuncheon Sacred Heart Hospital, Hallym University College of Medicine, Chuncheon-si 24253, Republic of Korea 2. Institute of New Frontier Research Team, Hallym University College of Medicine, Chuncheon-si 24252, Republic of Korea 3. Department of Anesthesiology and Pain Medicine, College of Medicine, Kangwon National University, Chuncheon-si 24341, Republic of Korea
Abstract
Postoperative pulmonary edema (PPE) is a well-known postoperative complication. We hypothesized that a machine learning model could predict PPE risk using pre- and intraoperative data, thereby improving postoperative management. This retrospective study analyzed the medical records of patients aged > 18 years who underwent surgery between January 2011 and November 2021 at five South Korean hospitals. Data from four hospitals (n = 221,908) were used as the training dataset, whereas data from the remaining hospital (n = 34,991) were used as the test dataset. The machine learning algorithms used were extreme gradient boosting, light-gradient boosting machine, multilayer perceptron, logistic regression, and balanced random forest (BRF). The prediction abilities of the machine learning models were assessed using the area under the receiver operating characteristic curve, feature importance, and average precisions of precision-recall curve, precision, recall, f1 score, and accuracy. PPE occurred in 3584 (1.6%) and 1896 (5.4%) patients in the training and test sets, respectively. The BRF model exhibited the best performance (area under the receiver operating characteristic curve: 0.91, 95% confidence interval: 0.84–0.98). However, its precision and f1 score metrics were not good. The five major features included arterial line monitoring, American Society of Anesthesiologists physical status, urine output, age, and Foley catheter status. Machine learning models (e.g., BRF) could predict PPE risk and improve clinical decision-making, thereby enhancing postoperative management.
Funder
Hallym University Korea Health Industry Development Institute
Reference41 articles.
1. Diagnosis, prevention and management of postoperative pulmonary edema;Bajwa;Ann. Med. Health Sci. Res.,2012 2. Cardiomyopathy of overload. A major determinant of prognosis in congestive heart failure;Katz;N. Engl. J. Med.,1990 3. Kwon, Y.-S., Kim, H., Lee, H., Kim, J.-H., Jang, J.-S., Hwang, S.-M., Hong, J.-Y., Yang, G.-E., Kim, Y., and Lee, J.-J. (2021). Effect of Intra- and Post-Operative Fluid and Blood Volume on Postoperative Pulmonary Edema in Patients with Intraoperative Massive Bleeding. J. Clin. Med., 10. 4. Neurogenic pulmonary edema;Baumann;Acta Anaesthesiol. Scand.,2007 5. Matthew Wemple, M.H., Andrew, M.M.D., and Luks, M.D. (2022, October 24). Neurogenic Pulmonary Edema. Available online: https://www.uptodate.com/contents/neurogenic-pulmonary-edema.
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献
|
|