A natural language processing algorithm identified patients with atrial fibrillation using clinical free text with an F-score of 94.2% at site 1 and 92.5% at site 2 (P<0.001).
Observational (n=1,447)
Sí
Does a natural language processing algorithm accurately identify patients with atrial fibrillation using clinical notes compared to clinician adjudication?
A natural language processing algorithm can accurately identify patients with atrial fibrillation from free-text clinical notes across different sites, facilitating high-throughput cohort identification.
valor p: p=<0.001
Background: The electronic medical record contains a wealth of information buried in free text. We created a natural language processing algorithm to identify patients with atrial fibrillation (AF) using text alone. Methods and Results: We created 3 data sets from patients with at least one AF billing code from 2010 to 2017: a training set (n=886), an internal validation set from site no. 1 (n=285), and an external validation set from site no. 2 (n=276). A team of clinicians reviewed and adjudicated patients as AF present or absent, which served as the reference standard. We trained 54 algorithms to classify each patient, varying the model, number of features, number of stop words, and the method used to create the feature set. The algorithm with the highest F-score (the harmonic mean of sensitivity and positive predictive value) in the training set was applied to the validation sets. F-scores and area under the receiver operating characteristic curves were compared between site no. 1 and site no. 2 using bootstrapping. Adjudicated AF prevalence was 75.1% at site no. 1 and 86.2% at site no. 2. Among 54 algorithms, the best performing model was logistic regression, using 1000 features, 100 stop words, and term frequency-inverse document frequency method to create the feature set, with sensitivity 92.8%, specificity 93.9%, and an area under the receiver operating characteristic curve of 0.93 in the training set. The performance at site no. 1 was sensitivity 92.5%, specificity 88.7%, with an area under the receiver operating characteristic curve of 0.91. The performance at site no. 2 was sensitivity 89.5%, specificity 71.1%, with an area under the receiver operating characteristic curve of 0.80. The F-score was lower at site no. 2 compared with site no. 1 (92.5% SD, 1.1% versus 94.2% SD, 1.1%; P 90% F-score at 2 separate sites. This approach allows better use of the clinical narrative and creates an opportunity for precise, high-throughput cohort identification.
Shah et al. (Thu,) conducted a observational in Atrial fibrillation (n=1,447). Natural language processing algorithm vs. Clinician adjudication (reference standard) was evaluated on F-score and area under the receiver operating characteristic curve (p=<0.001). A natural language processing algorithm identified patients with atrial fibrillation using clinical free text with an F-score of 94.2% at site 1 and 92.5% at site 2 (P<0.001).