The article explores the development of a hybrid approach combining rule-based and machine learning methods to effectively detect social support and social isolation within clinical notes from electronic health records (EHRs). Given the time-intensive nature of manual data extraction, the research aims to annotate key terms that delineate an individual's social state. Inter-annotator agreement (IAA) is crucial in this context, with a systematic process established for refining annotation guidelines to ensure robust analytic performance and practical clinical applications.
Social isolation (SI) reflects a state of limited contact with community, while social support (SS) illustrates the comfort gained from social connections.
The hybrid approach utilizes rule-based and machine learning techniques in natural language processing to effectively identify social support and isolation in clinical notes.
The annotation process involves a rigorous, iterative approach for achieving high inter-annotator agreement before establishing a gold standard for clinical notes analysis.
By advancing methods to extract social data from electronic health records, this research addresses the critical need for efficient analysis of patient social health.
#social-support #social-isolation #natural-language-processing #electronic-health-records #machine-learning
Collection
[
|
...
]