Institute for Medical Informatics, Statistics and Epidemiology (IMISE), Leipzig University, Leipzig, Germany.
Stud Health Technol Inform. 2024 Aug 30;317:190-199. doi: 10.3233/SHTI240855.
Medical terminologies and code systems, which play a vital role in the health domain, are rarely static but undergo changes as knowledge and terminology evolves. This includes addition, deletion and relabeling of terms, and, if terms are organized hierarchically, changing their position. Tracking these changes may become important if one uses multiple versions of the same terminology and interoperability is desired.
We propose a new method for automatic change tracking between terminology versions. It consists of a declarative import pipeline, which translates source terminologies into a common data model. We then use semantic and lexical change detection algorithms. They produce an ontology-based representation of terminology changes, which can be queried using semantic query languages.
The method proves accurate in detecting additions, deletions, relocations and renaming of terms. In cases where inter-version term mapping information is provided by the publisher, we were able to highly enhance the ability to differentiate between simple additions/deletions and refinements/consolidation of terms.
The method proves effective for semi-automatic change handling if term refinements and consolidation are relevant and for automatic change detection if additional mapping information is available.
医学术语和编码系统在医疗领域中起着至关重要的作用，它们很少是静态的，而是随着知识和术语的发展而不断变化。这包括术语的添加、删除和重新标记，如果术语是按层次组织的，则还包括改变其位置。如果使用同一术语的多个版本并且需要互操作性，那么跟踪这些变化可能变得很重要。
我们提出了一种用于术语版本之间自动更改跟踪的新方法。它由一个声明性的导入管道组成，该管道将源术语转换为通用数据模型。然后，我们使用语义和词汇更改检测算法。它们生成基于本体的术语更改表示，可以使用语义查询语言对其进行查询。
该方法在检测术语的添加、删除、重新定位和重命名方面证明是准确的。在发布者提供跨版本术语映射信息的情况下，我们能够极大地提高区分简单添加/删除与术语细化和合并的能力。
如果术语细化和合并相关，则该方法对于半自动更改处理是有效的，如果有其他映射信息，则对于自动更改检测是有效的。
