Scholarlink

Background <p>Standardizing data from Traditional Korean Medicine (TKM) is essential for enhancing interoperability with international real-world data infrastructures, such as multi-institutional OMOP-CDM databases within the OHDSI network, multinational claims databases, and large-scale clinical data repositories, thereby enabling evidence-based research. This study aimed to convert electronic health records (EHRs) from a TKM hospital into the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM).</p> Methods <p>We transformed the EHR data from Wonkwang University Gwangju Korean Medicine Hospital into the OMOP CDM format. TKM-specific diagnoses, procedures, and medications were mapped to standard vocabularies, and new concept codes were developed when no standard terms were available. Beyond a direct application of conventional CDM transformation procedures, we established explicit mapping principles and introduced structural extensions-such as the KIOM terminology system-to accommodate pattern identification, complex herbal prescriptions, and detailed acupuncture techniques that could not be represented within existing vocabularies. An Extract, Transform, Load (ETL) process was conducted, and data quality was evaluated using the ACHILLES tool.</p> Results <p>The converted dataset, named the Wonkwang Traditional Korean Medicine (WKTKM) database, included records from 88,449 patients. It comprised more than 4&#xa0;million condition records, approximately 10&#xa0;million drug prescriptions, and over 10&#xa0;million procedure records. Most laboratory results and medications were successfully mapped to existing standard concepts, while TKM-specific concepts were integrated using a newly developed terminology system called the KIOM codes.</p> Conclusion <p>This proof-of-concept study demonstrates the technical feasibility of deterministic, rule-based transformation of EHR data from a TKM hospital into the OMOP CDM. By incorporating methodological extensions-including the temporary creation of KIOM terminology and the development of generic mapping rules for TKM-specific diagnostic and therapeutic elements-this study illustrates a practical baseline pathway for harmonizing traditional medicine data. While the WKTKM database establishes an important methodological foundation, extensive cross-institutional validation and centralized terminology governance are necessary prerequisites before such standardized datasets can be reliably deployed for large-scale, generalizable international collaborative research.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Transforming Traditional Korean Medicine hospital EHRs into the OMOP common Data Model: methodology and implications

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月16日

查看全文
  • Man Young Park,
  • Jaeuk U. Kim,
  • SunMi Choi,
  • Youngheum Yoon,
  • Byung-Kwan Seo,
  • Sangkwan Lee

摘要

Background

Standardizing data from Traditional Korean Medicine (TKM) is essential for enhancing interoperability with international real-world data infrastructures, such as multi-institutional OMOP-CDM databases within the OHDSI network, multinational claims databases, and large-scale clinical data repositories, thereby enabling evidence-based research. This study aimed to convert electronic health records (EHRs) from a TKM hospital into the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM).

Methods

We transformed the EHR data from Wonkwang University Gwangju Korean Medicine Hospital into the OMOP CDM format. TKM-specific diagnoses, procedures, and medications were mapped to standard vocabularies, and new concept codes were developed when no standard terms were available. Beyond a direct application of conventional CDM transformation procedures, we established explicit mapping principles and introduced structural extensions-such as the KIOM terminology system-to accommodate pattern identification, complex herbal prescriptions, and detailed acupuncture techniques that could not be represented within existing vocabularies. An Extract, Transform, Load (ETL) process was conducted, and data quality was evaluated using the ACHILLES tool.

Results

The converted dataset, named the Wonkwang Traditional Korean Medicine (WKTKM) database, included records from 88,449 patients. It comprised more than 4 million condition records, approximately 10 million drug prescriptions, and over 10 million procedure records. Most laboratory results and medications were successfully mapped to existing standard concepts, while TKM-specific concepts were integrated using a newly developed terminology system called the KIOM codes.

Conclusion

This proof-of-concept study demonstrates the technical feasibility of deterministic, rule-based transformation of EHR data from a TKM hospital into the OMOP CDM. By incorporating methodological extensions-including the temporary creation of KIOM terminology and the development of generic mapping rules for TKM-specific diagnostic and therapeutic elements-this study illustrates a practical baseline pathway for harmonizing traditional medicine data. While the WKTKM database establishes an important methodological foundation, extensive cross-institutional validation and centralized terminology governance are necessary prerequisites before such standardized datasets can be reliably deployed for large-scale, generalizable international collaborative research.