中医临床术语分类标准框架的思考与构建探索

在当前世界性的医疗信息标准化发展潮流中,中医临床术语标准化工作势在必行.中医临床术语分类标准框架是中医临床术语标准化的基础,在国内乃至国际上传统医学领域中均无先例和具体的方法指导.本研究通过调研一体化语言系统、SNOMED、ICD等国外分类代码标准,以及我国的中医药信息标准化的相关工作,通过研究分析中医学与现代医学的理论体系特点和临床诊疗思路的异同点,借鉴SNOMED CT的构建模式,以中医药学自身独特理论及学科特点为核心,探索构建中医临床术语分类标准框架.

SNOMED发展概况与展望

SNOMED(the Systemized Nomenclafure of Medicine)[1]是由美国病理学家学会(college of American Pathologist,CAP)[2]负责牵头开发并经过科学验证的一部临床卫生保健术语集,它的应用可使卫生保健知识更加易于获得并应用于实践,为临床信息数据的采集、聚合处理和共享提供便利[3],其先进的工作模式和管理经验值得我们学习和借鉴.

SNOMEDCT层次结构与标识符

SNOMED CT中的概念都通过层级关系相连.介绍了目前SNOMD CT的概念层级结构及分类情况,包括新增加的"元数据层级"、"根元数据概念";还介绍了常用的几个标识符及SNOMED CT标识符(SCTID)的结构,包括数据类型、扩展、约束条件、验校码、分区标识符、命名空间等,为学习和借鉴SNOMD CT的标准化方法提供资料.

SNOMEDCT层次结构与标识符

SNOMED CT中的概念都通过层级关系相连.介绍了目前SNOMD CT的概念层级结构及分类情况,包括新增加的"元数据层级"、"根元数据概念";还介绍了常用的几个标识符及SNOMED CT标识符(SCTID)的结构,包括数据类型、扩展、约束条件、验校码、分区标识符、命名空间等,为学习和借鉴SNOMD CT的标准化方法提供资料.

中医临床术语编码方式研究

目前大部分医院都已配置了HIS、PACS、LIS及CIS，由于各系统为不同目的建立，系统间存在集成性低、异构性突出等问题，使得系统采集的数据不能互联互通。要实现医疗机构信息的交互，需要共享标准规范的可控术语词表，合理的术语编码方案可以更好实现自然语言与机器语言的统一和转化。术语编码是计算机识别和定位术语的唯一凭借，通过对临床数据进行标准化编码可以对庞大的数据资源进行有效管理和利用，有利于数据的交换、共享和再分析。基于国内国际术语集中术语编码方式的研究，依据SNOMED CT编码方式，提出适合信息系统的中医临床术语编码方案。

The Theory of Bio-Medical Knowledge Integration(Ⅷ)——The Intrinsic Medical Informatics(IMI) and Meta-dimensions(MDs) of SNOMED

This paper presented the profile of Intrinsic Medical Informatics(IMI), which consists of Medical Cognition Informatics(MCI) and Organism Informatics (OI). MCI and OI are taken as two cornerstones of the Theory of BioMedical Knowledge Integration(BMKI). Additionally, so called Meta-dimension architecture of SNOMED is discussed in order to venture its context computing.

From the integrated view of cognitive and noetic sciences, a set of meta-dimensions and their combinations, covering almost all of general linkage and modifiers of SNOMED, has been presented. Some examples for reasoning and operating for this set have been illustrated. Therefore the article has proved theoretically that it is possible to significantly increase the abilities of reasoning or operating and being-understood by computer in data mining of EMR, etc for a terminology system including relation(or linkage) or restriction (or modifier) items such as SNOMED. It is believed also the realizations of these tasks expected by this article are not difficult in technologic aspects for the modern software and data base technologies and the good IT engineers.