Application of machine learning and natural language processing for predicting stroke-associated pneumonia

Abstract

BackgroundIdentifying patients at high risk of stroke-associated pneumonia (SAP) may permit targeting potential interventions to reduce its incidence. We aimed to explore the functionality of machine learning (ML) and natural language processing techniques on structured data and unstructured clinical text to predict SAP by comparing it to conventional risk scores.MethodsLinked data between a hospital stroke registry and a deidentified research-based database including electronic health records and administrative claims data was used. Natural language processing was applied to extract textual features from clinical notes. The random forest algorithm was used to build ML models. The predictive performance of ML models was compared with the A2DS2, ISAN, PNA, and ACDD4 scores using the area under the receiver operating characteristic curve (AUC).ResultsAmong 5,913 acute stroke patients hospitalized between Oct 2010 and Sep 2021, 450 (7.6%) developed SAP within the first 7 days after stroke onset. The ML model based on both textual features and structured variables had the highest AUC [0.840, 95% confidence interval (CI) 0.806–0.875], significantly higher than those of the ML model based on structured variables alone (0.828, 95% CI 0.793–0.863, P = 0.040), ACDD4 (0.807, 95% CI 0.766–0.849, P = 0.041), A2DS2 (0.803, 95% CI 0.762–0.845, P = 0.013), ISAN (0.795, 95% CI 0.752–0.837, P = 0.009), and PNA (0.778, 95% CI 0.735–0.822, P < 0.001). All models demonstrated adequate calibration except for the A2DS2 score.ConclusionsThe ML model based on both textural features and structured variables performed better than conventional risk scores in predicting SAP. The workflow used to generate ML prediction models can be disseminated for local adaptation by individual healthcare organizations.

Ventilator-Associated Pneumonia Prediction Models Based on AI: Scoping Review

“…Furthermore, much data remain to be mined. Tsai et al [ 18 ] found that information extracted from unstructured clinical text could make predictive models more comprehensive and improve their predictive performance. In addition to unstructured clinical text, lung radiography and computed tomography can be used to predict the occurrence of pneumonia.…”

Characteristics and risk of stroke in emergency department patients with acute dizziness

“…While previous research has explored the immediate risk of stroke in dizzy patients, this study uniquely investigated long-term stroke risk and types of strokes post-ED discharge. Moreover, this study utilized data from the DRD, a hospital research-based database containing long-term and varied healthcare data for a large patient population [ 18 ]. The comprehensive nature of the DRD further enhanced the study's capability to capture and analyze relevant outcomes.…”

“…The study data were extracted from the Ditmanson Research Database (DRD), a deidentified research-based database containing administrative claims data, electronic medical records, and vital status information provided by the National Death Index. The details of the DRD have been described elsewhere [ 18 ]. Briefly, the DRD holds clinical information for approximately 1.6 million patients who visited the study hospital.…”

Machine learning using multimodal and autonomic nervous system parameters predicts clinically apparent stroke-associated pneumonia in a development and testing study

“…Another comparative study concluded that the clinical prediction scores varied in their simplicity of use and, while comparable in performance, their utility for preventive intervention trials and in clinical practice required further investigation [ 21 ]. More recently, ML-based prediction of SAP, including methods based on natural language processing, has also been explored [ 17 , 18 ]. These studies reported AUCs of 0.84 which is below the AUC of 0.91 reported here.…”