2023 Fiscal Year Final Research Report

Medical Terminology Analysis: Word Structure and Semantic Classification of Compound Words to Support Linguistic Processing of Real Data

Research Project

PDF

Project/Area Number	21H03777
Research Category	Grant-in-Aid for Scientific Research (B)
Allocation Type	Single-year Grants
Section	一般
Review Section	Basic Section 90020:Library and information science, humanistic and social informatics-related
Research Institution	Nara Institute of Science and Technology (2023) Seinan Jo Gakuin University (2021-2022)
Principal Investigator	Sagara Kaoru 奈良先端科学技術大学院大学, 先端科学技術研究科, 客員准教授 (00330887)
Co-Investigator(Kenkyū-buntansha)	東条佳奈大阪大学, 大学院人文学研究科(人文学専攻、芸術学専攻、日本学専攻), 講師 (20782220) 山崎誠大学共同利用機関法人人間文化研究機構国立国語研究所, 研究系, 客員教授 (30182489) 黒田航杏林大学, 医学部, 准教授 (30425764) 麻子軒関西大学, 国際教育センター, 留学生別科特任常勤講師 (30880249) 高崎智子西南女学院大学, 保健福祉学部, 教授 (30882865)
Project Period (FY)	2021-04-01 – 2024-03-31
Keywords	医療用語 / 合成語 / 語構造 / 意味分類 / 電子カルテ
Outline of Final Research Achievements	The 810 compound words in the medical records were classified into three categories: those listed in the glossary and used idiomatically, those that are not idiomatic and those that are considered to correspond to a series of sentences or phrases without functional words. The 810 compound words were then segmented from a medical perspective and the resulting 1,264 word components were assigned 50 different semantic labels. Furthermore, a method for describing the word structure of compound words was defined. A lexical table consisting of these was then created. Furthermore, a support tool with pre-processing functions, including normalization, word segmentation and the extraction of compound words and word components, was created to enable healthcare professionals to process the accumulated medical record data themselves in their facilities.
Free Research Field	医療言語処理
Academic Significance and Societal Importance of the Research Achievements	医療記録に含まれる合成語には、辞書の見出し語にはない「食渣停滞」など、単語の一般的な意味から全体の意味の推測が困難なものがある。これらを医療の観点から語分割し、各要素に意味ラベルを付与したことで、言い換えが可能となり、機械学習や医療従事者向けの教育に活用できる。また、表記の揺れや同義語の統制が可能となることから、自然言語処理の精度の向上が期待できる。加えて本言語処理支援ツールは、医療従事者が施設内で個人情報を含む医療記録の語分割および用語抽出を可能にする。