Prediction of 30-Day Readmission After Stroke Using Machine Learning and Natural Language Processing

Abstract

Background and Purpose: This study aims to determine whether machine learning (ML) and natural language processing (NLP) from electronic health records (EHR) improve the prediction of 30-day readmission after stroke.Methods: Among index stroke admissions between 2011 and 2016 at an academic medical center, we abstracted discrete data from the EHR on demographics, risk factors, medications, hospital complications, and discharge destination and unstructured textual data from clinician notes. Readmission was defined as any unplanned hospital admission within 30 days of discharge. We developed models to predict two separate outcomes, as follows: (1) 30-day all-cause readmission and (2) 30-day stroke readmission. We compared the performance of logistic regression with advanced ML algorithms. We used several NLP methods to generate additional features from unstructured textual reports. We evaluated the performance of prediction models using a five-fold validation and tested the best model in a held-out test dataset. Areas under the curve (AUCs) were used to compare discrimination of each model.Results: In a held-out test dataset, advanced ML methods along with NLP features out performed logistic regression for all-cause readmission (AUC, 0.64 vs. 0.58; p < 0.001) and stroke readmission prediction (AUC, 0.62 vs. 0.52; p < 0.001).Conclusion: NLP-enhanced machine learning models potentially advance our ability to predict readmission after stroke. However, further improvement is necessary before being implemented in clinical practice given the weak discrimination.

Natural Language Processing for Smart Healthcare

“…Before the interpretability issue is fully explored, the role of decision support systems in clinical practice can only be auxiliary from the perspectives of medical ethics and practical application. [133], [134] machine translation clinical documentation [135]- [137] speech recognition clinical decision support build QA-based clinical decision support systems [88], [138], [139] information extraction build clinical decision support systems with extracted information: family history information [140], entities and relations [141], [142], treatment and prognosis data [143], clinical data concepts and features [144], causal relations [113], [114] question answering healthcare quality control: assess clinical procedures [145], [146], warning of ADE [147], disease symptoms [148], [149], and outcome-related causal effects [150] information extraction, causal inference provide supporting evidence for decisions under evidence-based fashion [112], [151]- [154] information retrieval, causal inference Hospital Management medical resource allocation patient triage [155], [156] information extraction enable users to communicate and control intelligent systems through virtual assistants [157]- [159], hospital automation systems [160], [161] and collaborative robots [162], [163] speech recognition, natural language understanding predict and reduce readmission rate [164]- [166] information extraction free medical staff from routine text writing [74], [167] information extraction data management m...…”

“…Virtual assistants [157]- [159], hospital automation systems [160], [161] and collaborative robots [162], [163] with voice control can further reduce the burden on medical staff, thereby improving hospital management efficiency. There are also some interesting works that have explored the prediction of patient readmission to rearrange medical resources/interventions and reduce the readmission rate [164]- [166]. In addition, by leveraging text generation techniques, part of text writing in healthcare, especially routine reports, can be taken over by machines, freeing medical staff from many administrative duties and making them available for direct patient care [74], [167].…”

The Allure of Big Data to Improve Stroke Outcomes: Review of Current Literature

“…Models involving natural language processing of unstructured free-text magnetic resonance imaging reports have been successfully used to predict 90-day functional outcomes with great accuracy [63]. Machine learning methods have also been used to predict other stroke outcomes, including 30-day readmissions [60,64], home-time [50], motor recovery [65], and complications of stroke (e.g., pneumonia [66], and dysphagia [67]) and for developing a proxy measure of stroke severity [68]. Also, machine learning methods have been used to maximize the utility of longitudinal measures collected after acute stroke for improving the prediction of outcomes [65,66,68,69].…”

Predicting Hospital Readmission for Campylobacteriosis from Electronic Health Records: A Machine Learning and Text Mining Perspective

“…Risk factors for campylobacteriosis are widely recognized [5,7,36,37] but reliable predictors of hospitalisation have not been clearly established. Recently natural language processing techniques were adopted to predicit hospitalizations with structural and un-structural data [38,39]. Here, we develop a robust, validated, and cost-effective data-driven method to identify the most informative hospital readmission predictors from primary care medical records.…”