Extracting Modifiable Risk Factors from Narrative Preventive Healthcare Guidelines for EHR Integration

Shah, Setu; Luo, Xiao

doi:10.1109/bibe.2017.000-2

Cited by 2 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Specifically with regard to risk assessment, taking into account the textual context of a mention (other than negation) is relatively rare. Expressions such as “risk increase” or “60% risk” can be used for extracting factors affecting the likelihood of a disease from healthcare guidelines 19 or medical literature, 20 , 21 whereas understanding temporal aspects and linking laboratory tests with numerical results are necessary to recognize mentions of prespecified risk factors for cardiac artery disease (2014 i2b2 shared task) 22 or symptoms of appendicitis 23 …”

Section: Background and Significancementioning

confidence: 99%

Quantifying risk factors in medical reports with a context-aware linear model

Przybyła

Brockmeier

Ananiadou

2019

Journal of the American Medical Informatics Association

View full text Add to dashboard Cite

Objective We seek to quantify the mortality risk associated with mentions of medical concepts in textual electronic health records (EHRs). Recognizing mentions of named entities of relevant types (eg, conditions, symptoms, laboratory tests or behaviors) in text is a well-researched task. However, determining the level of risk associated with them is partly dependent on the textual context in which they appear, which may describe severity, temporal aspects, quantity, etc. Methods To take into account that a given word appearing in the context of different risk factors (medical concepts) can make different contributions toward risk level, we propose a multitask approach, called context-aware linear modeling, which can be applied using appropriately regularized linear regression. To improve the performance for risk factors unseen in training data (eg, rare diseases), we take into account their distributional similarity to other concepts. Results The evaluation is based on a corpus of 531 reports from EHRs with 99 376 risk factors rated manually by experts. While context-aware linear modeling significantly outperforms single-task models, taking into account concept similarity further improves performance, reaching the level of human annotators’ agreements. Conclusion Our results show that automatic quantification of risk factors in EHRs can achieve performance comparable to human assessment, and taking into account the multitask structure of the problem and the ability to handle rare concepts is crucial for its accuracy.

show abstract