Automated risk assessment of newly detected atrial fibrillation poststroke from electronic health record data using machine learning and natural language processing.

Background: Timely detection of atrial fibrillation (AF) after stroke is highly clinically relevant, aiding decisions on the optimal strategies for secondary prevention of stroke. In the context of limited medical resources, it is crucial to set the right priorities of extended heart rhythm monitoring by stratifying patients into different risk groups likely to have newly detected AF (NDAF). This study aimed to develop an electronic health record (EHR)-based machine learning model to assess the risk of NDAF in an early stage after stroke. Methods: Linked data between a hospital stroke registry and a deidentified research-based database including EHRs and administrative claims data was used. Demographic features, physiological measurements, routine laboratory results, and clinical free text were extracted from EHRs. The extreme gradient boosting algorithm was used to build the prediction model. The prediction performance was evaluated by the C-index and was compared to that of the AS5F and CHASE-LESS scores. Results: The study population consisted of a training set of 4,064 and a temporal test set of 1,492 patients. During a median follow-up of 10.2 months, the incidence rate of NDAF was 87.0 per 1,000 person-year in the test set. On the test set, the model based on both structured and unstructured data achieved a C-index of 0.840, which was significantly higher than those of the AS5F (0.779, p = 0.023) and CHASE-LESS (0.768, p = 0.005) scores. Conclusions: It is feasible to build a machine learning model to assess the risk of NDAF based on EHR data available at the time of hospital admission. Inclusion of information derived from clinical free text can significantly improve the model performance and may outperform risk scores developed using traditional statistical methods. Further studies are needed to assess the clinical usefulness of the prediction model.

Validation of ICD-9-CM and ICD-10-CM Diagnostic Codes for Identifying Patients with Out-of-Hospital Cardiac Arrest in a National Health Insurance Claims Database.

Purpose: Taiwan's national health insurance (NHI) database is a valuable resource for large-scale epidemiological and long-term survival research for out-of-hospital cardiac arrest (OHCA). We developed and validated case definition algorithms for OHCA based on the International Classification of Diseases (ICD) diagnostic codes and billing codes for NHI reimbursement. Patients and Methods: Claims data and medical records of all emergency department visits from 2010 to 2020 were retrieved from the hospital's research-based database. Death-related diagnostic codes and keywords were used to identify potential OHCA cases, which were ascertained by chart reviews. We tested the performance of the developed OHCA algorithms and validated them on an external dataset. Results: The algorithm defining OHCA as any cardiac arrest (CA)-related ICD code in the first three diagnosis fields performed the best with a sensitivity of 89.5% (95% confidence interval [CI], 88.2-90.7%), a positive predictive value (PPV) of 90.6% (95% CI, 89.4-91.8%), and a kappa value of 0.900 (95% CI, 0.891-0.909). The second-best algorithm consists of any CA-related ICD code in any diagnosis field with a billing code for triage acuity level 1, achieving a sensitivity of 85.6% (95% CI, 84.1-87.0%), a PPV of 93.6% (95% CI, 92.5-94.5), and a kappa value of 0.894 (95% CI, 0.884-0.903). Both algorithms performed well in external validation. In subgroup analyses, the former algorithm performed the best in adult patients, outpatient claims, and during the ICD-9 era. The latter algorithm performed the best in the inpatient claims and during the ICD-10 era. The best algorithm for identifying pediatric OHCAs was any CA-related ICD code in the first three diagnosis fields with a billing code for triage acuity level 1. Conclusion: Our results may serve as a reference for future OHCA studies using the Taiwan NHI database.

Natural Language Processing Enhances Prediction of Functional Outcome After Acute Ischemic Stroke.

Background Conventional prognostic scores usually require predefined clinical variables to predict outcome. The advancement of natural language processing has made it feasible to derive meaning from unstructured data. We aimed to test whether using unstructured text in electronic health records can improve the prediction of functional outcome after acute ischemic stroke. Methods and Results Patients hospitalized for acute ischemic stroke were identified from 2 hospital stroke registries (3847 and 2668 patients, respectively). Prediction models developed using the first cohort were externally validated using the second cohort, and vice versa. Free text in the history of present illness and computed tomography reports was used to build machine learning models using natural language processing to predict poor functional outcome at 90 days poststroke. Four conventional prognostic models were used as baseline models. The area under the receiver operating characteristic curves of the model using history of present illness in the internal and external validation sets were 0.820 and 0.792, respectively, which were comparable to the National Institutes of Health Stroke Scale score (0.811 and 0.807). The model using computed tomography reports achieved area under the receiver operating characteristic curves of 0.758 and 0.658. Adding information from clinical text significantly improved the predictive performance of each baseline model in terms of area under the receiver operating characteristic curves, net reclassification improvement, and integrated discrimination improvement indices (all P<0.001). Swapping the study cohorts led to similar results. Conclusions By using natural language processing, unstructured text in electronic health records can provide an alternative tool for stroke prognostication, and even enhance the performance of existing prognostic scores.