A clinical use case to evaluate the i2b2 Hive: predicting asthma exacerbations

• Published in: AMIA ... Annual Symposium proceedings Vol. 2009; pp. 442 - 446
• Main Authors: Meystre, Stéphane M, Deshmukh, Vikrant G, Mitchell, Joyce
• Format: Journal Article
• Language: English
• Published: United States American Medical Informatics Association 14.11.2009
• Subjects: Adult; Asthma - diagnosis; Female; Humans; Informatics; Information Storage and Retrieval - methods; Male; Medical Records Systems, Computerized; Natural Language Processing; Prognosis; Software
• ISSN: 1559-4076

To evaluate the i2b2 Hive as a tool to query, visualize, and extract clinical data, we selected a use case from the i2b2 airways diseases driving biology project: asthma exacerbations prediction. We analyzed the cohort selection and the extraction of the clinical data used by this asthma exacerbations prediction study. The structured data included the asthma diagnosis, birthdate, age, race, sex, height, weight, and BMI. The smoking status is typically only mentioned in clinical notes, and we evaluated the Natural Language Processing (NLP) application embedded in the i2b2 NLP cell to extract the smoking status from history and physical exam reports.Querying structured data was possible with the i2b2 workbench for about half the clinical data elements. The remaining had to be queried using a commercial database management system client. The automated extraction of the smoking status reached a mean precision of 0.79 and a mean specificity of 0.90.