Using local lexicalized rules to identify heart disease risk factors in clinical notes

• Published in: Journal of biomedical informatics 2015-12, Vol.58 (Suppl), p.S183-S188
• Main Authors: Karystianis, George, Dehghan, Azad, Kovacevic, Aleksandar, Keane, John A., Nenadic, Goran
• Format: Article
• Language: English
• Subjects: Aged; Cardiovascular Diseases - diagnosis; Cardiovascular Diseases - epidemiology; Cohort Studies; Comorbidity; Computer Security; Confidentiality; Coronary artery disease; Data Mining - methods; Diabetes; Diabetes Complications - diagnosis; Diabetes Complications - epidemiology; Electronic Health Records - organization & administration; Female; Heart disease; Heart diseases; Humans; Hypertension; Incidence; Longitudinal Studies; Male; Methodology; Middle Aged; Narration; Natural Language Processing; Pattern Recognition, Automated - methods; Recall; Risk analysis; Risk Assessment - methods; Risk factors; Rule-based modelling; Semantics; Tasks; Text mining; United Kingdom - epidemiology; Vocabularies; Vocabulary, Controlled
• ISSN: 1532-0464; 1532-0480
• DOI: 10.1016/j.jbi.2015.06.013

[Display omitted] •We created a set of task-specific dictionaries related to heart disease.•We designed generic rules for risk factors identification.•The results are aggregated at the document level.•Temporal attributes are assigned class-specific defaults.•Rule-based risk factor extraction is feasible and reliable. Heart disease is the leading cause of death globally and a significant part of the human population lives with it. A number of risk factors have been recognized as contributing to the disease, including obesity, coronary artery disease (CAD), hypertension, hyperlipidemia, diabetes, smoking, and family history of premature CAD. This paper describes and evaluates a methodology to extract mentions of such risk factors from diabetic clinical notes, which was a task of the i2b2/UTHealth 2014 Challenge in Natural Language Processing for Clinical Data. The methodology is knowledge-driven and the system implements local lexicalized rules (based on syntactical patterns observed in notes) combined with manually constructed dictionaries that characterize the domain. A part of the task was also to detect the time interval in which the risk factors were present in a patient. The system was applied to an evaluation set of 514 unseen notes and achieved a micro-average F-score of 88% (with 86% precision and 90% recall). While the identification of CAD family history, medication and some of the related disease factors (e.g. hypertension, diabetes, hyperlipidemia) showed quite good results, the identification of CAD-specific indicators proved to be more challenging (F-score of 74%). Overall, the results are encouraging and suggested that automated text mining methods can be used to process clinical notes to identify risk factors and monitor progression of heart disease on a large-scale, providing necessary data for clinical and epidemiological studies.