ODD: A Benchmark Dataset for the NLP-based Opioid Related Aberrant Behavior Detection

arXiv:2307.02591v3 Announce Type: replace
Abstract: Opioid related aberrant behaviors (ORABs) present novel risk factors for opioid overdose. This paper introduces a novel biomedical natural language processing benchmark dataset named ODD, for ORAB Detection Dataset. ODD is an expert-annotated dataset designed to identify ORABs from patients' EHR notes and classify them into nine categories; 1) Confirmed Aberrant Behavior, 2) Suggested Aberrant Behavior, 3) Opioids, 4) Indication, 5) Diagnosed opioid dependency, 6) Benzodiazepines, 7) Medication Changes, 8) Central Nervous System-related, and 9) …

abstract arxiv behavior behavior detection benchmark biomedical cs.ai cs.cl dataset detection ehr expert identify language language processing natural natural language natural language processing nlp notes novel opioid paper patients processing risk type