Using machine learning to predict stroke-associated pneumonia in Chinese patients with acute ischemic stroke.

BACKGROUND AND PURPOSE: Stroke-associated pneumonia is a common, severe but preventable complication after acute ischemic stroke (AIS). Early identification of patients at high risk of stroke-associated pneumonia is especially necessary. However, the previous prediction models were not widely used in clinical practice. Thus, we aimed to develop a model to predict stroke-associated pneumonia in Chinese AIS patients using machine learning methods. METHODS: AIS patients were prospectively collected at the National Advanced Stroke Center of Nanjing First Hospital (China) between September 2016 and November 2019, and the data was randomly subdivided into a training set and a testing set. With the training set, five machine learning models (logistic regression with regulation, support vector machine, random forest classifier, extreme gradient boosting and fully-connected deep neural network) were developed. These models were assessed by the area under curve of receiver operating characteristic on the testing set. Our models were also compared with ISAN score and PNA score. RESULTS: 3160 AIS patients were eventually included into this retrospective study. Among the five machine learning models, extreme gradient boosting model performed best. The area under curve of the extreme gradient boosting model on the testing set was 0.841 (sensitivity: 81.0%; specificity: 73.3%). It also achieved significantly better performance than ISAN score and PNA score. CONCLUSIONS: Our study firstly demonstrated that the extreme gradient boosting model with six common variables can predict stroke-associated pneumonia in Chinese AIS patients more optimally than ISAN score and PNA score.