Natural language processing (NLP) is a computer technology that uses language cues and context to interpret findings and mine large databases to identify desired clinical information. The aim of this study was to develop and validate an NLP algorithm to identify dysplasia in patients with Barrett’s esophagus (BE) using histopathology reports in a large integrated electronic medical record system.
In this study, investigators used national Veterans Affairs databases to identify patients with suspected BE; 600 pathology reports were randomly selected for NLP development and 400 for validation. The NLP algorithm identified dysplasia with 98.0% accuracy, 91.7% recall, and 93.2% precision, with an F-measure of 92.4% in the development set of 600 patients. Among the 400 patients in the validation cohort, the NLP algorithm identified dysplasia with 98.7% accuracy, 92.3% recall, and 100.0% precision, with an F-measure of 96.0%, compared with manual review. The algorithm correctly classified all 7 patients with confirmed high-grade dysplasia.
Nguyen Wenker T, Natarajan Y, Caskey K, et al. Using natural language processing to automatically identify dysplasia in pathology reports for patients with Barrett's esophagus. Clin Gastroenterol Hepatol