Harnessing Machine Learning for Prediction of Postoperative Pulmonary Complications: Retrospective Cohort Design

Jae-Jun Lee,So-Young Lim,S. Hwang,Jong-Ho Kim,Boreum Cheon,Youngsuk Kwon,Minguan Kim

DOI: https://doi.org/10.3390/jcm12175681

IF: 3.9

2023-08-31

Journal of Clinical Medicine

Abstract:Postoperative pulmonary complications (PPCs) are significant causes of postoperative morbidity and mortality. This study presents the utilization of machine learning for predicting PPCs and aims to identify the important features of the prediction models. This study used a retrospective cohort design and collected data from two hospitals. The dataset included perioperative variables such as patient characteristics, preexisting diseases, and intraoperative factors. Various algorithms, including logistic regression, random forest, light-gradient boosting machines, extreme-gradient boosting machines, and multilayer perceptrons, have been employed for model development and evaluation. This study enrolled 111,212 adult patients, with an overall incidence rate of 8.6% for developing PPCs. The area under the receiver-operating characteristic curve (AUROC) of the models was 0.699–0.767, and the f1 score was 0.446–0.526. In the prediction models, except for multilayer perceptron, the 10 most important features were obtained. In feature-reduced models, including 10 important features, the AUROC was 0.627–0.749, and the f1 score was 0.365–0.485. The number of packed red cells, urine, and rocuronium doses were similar in the three models. In conclusion, machine learning provides valuable insights into PPC prediction, significant features for prediction, and the feasibility of models that reduce the number of features.

Medicine,Computer Science

What problem does this paper attempt to address?

The paper attempts to address the problem of predicting postoperative pulmonary complications (PPCs) using machine learning techniques and identifying important features in the predictive model.

### Core Issues of the Paper:
1. **Predicting Postoperative Pulmonary Complications**: The study aims to develop machine learning models that can predict whether patients will experience pulmonary complications after surgery.
2. **Identifying Important Features**: The study seeks to identify the most important features related to postoperative pulmonary complications through machine learning models to improve prediction accuracy.

### Main Objectives:
- Use various machine learning algorithms (such as logistic regression, random forest, gradient boosting machine, etc.) to build predictive models.
- Analyze the importance of different features in the predictive models.
- Develop simplified models that include only the most important features and evaluate their performance.

### Data Sources and Methods:
- The study used data from two hospitals, including a total of 111,212 adult patients.
- The dataset included 102 preoperative and intraoperative variables.
- Five different machine learning algorithms were used for model training and evaluation.

### Results:
- The final models had AUC ranges between 0.699 and 0.767, and F1 scores between 0.446 and 0.526.
- In the simplified models, which included only the top 10 important features, performance decreased, but the random forest model performed the best.
- Important features in the models included the dose of rocuronium, urine output, and the amount of red blood cell transfusion.

### Discussion:
- Machine learning shows some potential in predicting postoperative pulmonary complications, but further improvements are needed to enhance prediction performance.
- Future research could consider incorporating more relevant data or using enhanced feature engineering methods to improve model performance.

### Conclusion:
This study demonstrates the application of machine learning in predicting postoperative pulmonary complications and identifies some key features, providing valuable references for clinical practice.