Relation Detection to Identify Stroke Assertions from Clinical Notes

Tech Stack: Python, Pandas, SciSpaCy, pyConText, Scikit-learn

Conference: MedInfo 2023 Notes: Sydney, Australia

Abstract: According to the World Stroke Organization, 12.2 million people worldwide will have their first stroke this year, almost half of which will die. Natural Language Processing (NLP) may improve stroke sub-phenotyping; however, existing rule-based classifiers are rigid, resulting in inadequate performance. This paper reports findings from a pilot study using NLP to enhance relation detection for stroke assertion detection to support research studies and healthcare operations.