Hua Xu, Ph.D.

A pervasive characteristic of clinical texts is their frequent use of abbreviations. Recognition and identification of abbreviations is an important, challenging task in clinical natural language processing (NLP). A comprehensive lexical resource comprised of all common, useful clinical abbreviations would have great applicability.

Abbreviation databases :

• LRABR from UMLS
• ADAM from Wei Zhou et al.
• 12,130 pathology abbreviations from Berman JJ.

Sense distribution database

• 448 abbreviations with sense frequency distribution from admission notes