AUTHOR=Jing Xiaolei , Wang Xueqi , Zhuang Hongxia , Fang Xiang , Xu Hao 

TITLE=Multiple Machine Learning Approaches Based on Postoperative Prediction of Pulmonary Complications in Patients With Emergency Cerebral Hemorrhage Surgery

JOURNAL=Frontiers in Surgery

VOLUME=Volume 8 - 2021

YEAR=2022

URL=https://www.frontiersin.org/journals/surgery/articles/10.3389/fsurg.2021.797872

DOI=10.3389/fsurg.2021.797872

ISSN=2296-875X

ABSTRACT=OBJECTIVE: This study aimed to create a prediction model of postoperative pulmonary complications for the patients with emergency cerebral hemorrhage surgery.
METHODS: Patients with hemorrhage surgery who underwent cerebral hemorrhage surgery were included and divided into 2 groups: patient with or without pulmonary complications. Patient characteristics, previous history, laboratory test and intervention were collected. Univariate and multivariate logistic regression were used to predict the postoperative pulmonary infection. Multiple machine learning approaches have been used to compare their importance in predicting factors, namely K-nearest neighbor(KNN), stochastic gradient descent (SGD), support vector classification (SVC), random forest (RF), and logistics regression(LR), as they are the most successful and widely used models for clinical data.
RESULTS: 354 patients with emergency cerebral hemorrhage surgery between January 1, 2017, and December 31, 2020 were included in the study. 53.7%(190/354) of the patients developed PPC. Stepwise logistic regression analysis revealed 4 independent predictive factors associated with pulmonary complications, including current smoker, lymphocyte count, clotting time and ASA score. In addition, the RF model had ideal predictive performance. 
CONCLUSIONS: According to our result, current smoker, lymphocyte count, clotting time and ASA score were independent risk of pulmonary complications. Machine learning approaches can also provide more evidence in the prediction of pulmonary complications.