This study evaluates the efficacy of a commercial medical Named Entity Recognition (NER) model combined with a post-processing protocol in identifying incidental pulmonary nodules from CT reports. We analyzed 9165 anonymized CT reports and classified them into 3 categories: no nodules, nodules present, and nodules >6 mm. For each report, a generic medical NER model annotated entities and their relations, which were then filtered through inclusion/exclusion criteria selected to identify pulmonary nodules.
View Article and Find Full Text PDFIn this article, we report on a 62-year-old non-cirrhotic male presenting to the emergency department (ED) with chronic abdominal pain, anorexia, and weight loss. Upon initial presentation, physical exam was unremarkable, other than for sarcopenia and splenomegaly. Initial imaging studies revealed a large thrombosis from the iliac vein to the right atrium of the heart.
View Article and Find Full Text PDFBackground: A variety of evidence-based algorithms and decision rules using D-Dimer testing have been proposed as instruments to allow physicians to safely rule out a pulmonary embolism (PE) in low-risk patients.
Objective: To describe the prevalence of D-Dimer utilization among emergency department (ED) physicians and its impact on positive yields and utilization rates of Computed Tomography Pulmonary Angiography (CTPA).
Methods: Data was collected on all CTPA studies ordered by ED physicians at three sites during a 2-year period.