Natural Language Processing Framework to Assess Clinical Conditions

Abstract

OBJECTIVE The authors developed a natural language processing (NLP) framework that could be used to extract clinical findings and diagnoses from dictated physician documentation. DESIGN De-identified documentation was made available by i2b2 Bio-informatics research group as a part of their NLP challenge focusing on obesity and its co-morbidities. The authors describe their approach, which used a combination of concept detection, context validation, and the application of a variety of rules to conclude patient diagnoses. RESULTS The framework was successful at correctly identifying diagnoses as judged by NLP challenge organizers when compared with a gold standard of physician annotations. The authors overall kappa values for agreement with the gold standard were 0.92 for explicit textual results and 0.91 for intuited results. The NLP framework compared favorably with those of the other entrants, placing third in textual results and fourth in intuited results in the i2b2 competition. CONCLUSIONS The framework and approach used to detect clinical conditions was reasonably successful at extracting 16 diagnoses related to obesity. The system and methodology merits further development, targeting clinically useful applications.

Development of Medical Information Systems (MISs)

“…They reported 97 % accuracy in predictions of disease status, which was comparable to that of humans. Ware and colleagues [ 325 ] also used the i2b2 NLP programs to focus on extracting diagnoses of obesity and 16 related diagnoses from textual dis-charge summary reports. They reported better than 90 % agreement with clinical experts as the comparative "gold standard."…”

Comparison of UMLS terminologies to identify risk of heart disease using clinical notes

“…The 2007 challenge [2] aimed at identification of obesity status of patients and associated co-morbidities using discharge summaries. Ware et al [10] developed a system where a principal concept of interest was identified using regular expressions and then surrounding context was searched for secondary concepts. Final decisions were made using these two sets of concepts.…”

Electronic Medical Records (EMRs), Epidemiology, and Epistemology: Reflections on EMRs and Future Pediatric Clinical Research

“…Natural language processing tools show considerable promise in converting free text into structured data,43 but these tools are far from perfected 44. In a recent unpublished analysis of free text for patient encounters for otitis media on a single day , a research group at the Children’s Hospital of Philadelphia encountered 278 different ways in 465 EMR notes to express the fact that patients had temperature > 102.0 F (e.g.…”