Yuimedi, a Medical Data Standardization Company, Provides Technical Support for NEC's OMOP Conversion Demonstration of NDB and Electronic Health Record Information

Yuimedi Inc. (Headquarters: Chuo-ku, Tokyo; Representative Director: Emiri Grimes; hereinafter "Yuimedi") announces that it provided technical support for OMOP CDM (hereinafter "OMOP") conversion in the "Research on OMOP CDM Conversion Technology Verification for Medical Information Standardization and Secondary Use," conducted jointly with NEC Corporation (Headquarters: Minato-ku, Tokyo; Director, Representative Executive Officer, President and CEO: Takayuki Morita; hereinafter "NEC"), Ehime University (Matsuyama City, Ehime Prefecture; University President: Hiroshige Nishina), and the Medical Data Linkage and Analysis Platform Association (hereinafter "FedAna").

For an overview of this verification, please refer to NEC's announcement. https://jpn.nec.com/press/202603/20260325_03.html

Background Expectations are rising for the utilization of public databases (hereinafter "public DBs"), including the National Database of Health Insurance Claims and Specific Health Checkups (NDB), for research and policy formulation, in line with the government's "National Medical Information Platform Initiative" and "Japan's EHDS." However, many public DBs are not structured for research purposes. The significant burden of data understanding and preprocessing has been a major obstacle to research utilization.

[Example in NDB analysis] Common Challenges | Details of Challenges | Solution Image with OMOP Utilization Complexity of Code Systems | Within a single domain, two problems are intertwined: the coexistence of code systems and the hierarchy of granularity. Taking pharmaceuticals as an example, multiple systems such as YJ codes, receipt computer codes, and HOT codes coexist for the same ingredient, and within each system, granularity is divided into ingredient, product, and dosage form levels. Therefore, to gather "patients prescribed statin drugs," dozens to hundreds of codes must be listed. | Multiple local codes are unified and mapped to OMOP standard vocabulary concept IDs. For statin drugs, regardless of brand-name, generic, or dosage form, data can be aggregated across the board with the same concept ID, significantly reducing code list management costs. Complex Period Calculation for Long-format Data | NDB records are generated per claim unit, so to understand the longitudinal history of "what treatment a patient received at which medical institution," data matching and complex cross-referencing processes are required. | The design philosophy is to aggregate data per patient, storing diagnoses, prescriptions, procedures, and tests chronologically under the same patient ID. Clinical events can be tracked and analyzed on a patient axis, allowing for natural analysis of treatment pathways and outcome evaluations. Analysis Protocols are Not Reusable | Analysis protocols created for NDB depend on NDB's unique data structure and cannot be reused for similar analyses on other databases. | With an OMOP-compliant DB, the same analysis protocol can be reused as is. Furthermore, analysis protocols developed and published by other countries for OMOP-compliant DBs can be brought into Japan, making participation in international collaborative research realistic.

Yuimedi's Role in This Verification In this project, Yuimedi led the technical aspects of high-quality conversion processing from synthetic NDB and electronic health record (HL7 FHIR) data prepared by NEC to OMOP. Specifically, Yuimedi conducted the formulation of conversion specifications, data conversion, quality checks, and the design of cohort definitions and visualization of features corresponding to research questions set for each type of public DB user. This demonstrated that OMOP-converted public DBs can be directly utilized for research.

The background to these achievements includes Yuimedi's accumulated research and development track record: * Yuimedi developed a conversion tool from HL7 FHIR to OMOP CDM in collaboration with Ehime University (2024/5/14) * Developed a new drug term mapping method utilizing LLM and RAG in collaboration with Ehime University (2025/3/26) * Evaluated OMOP concept conversion for all MEDIS standard disease name master data (2026/2/12)

Based on these achievements, Yuimedi continues to promote initiatives for medical data standardization in Japan.

Future Outlook The standardization of medical data is accelerating as a global trend. As governments and research institutions worldwide increasingly emphasize international data interoperability, OMOP is rapidly gaining popularity as a central standard, becoming a global standard. In Japan, momentum is also growing for the OMOP conversion of public DBs and electronic health record data, and this verification is a step to accelerate that trend.

Yuimedi, as a company specializing in OMOP-specific standardization technology at the core of its business, continues to work at the forefront of this field. As a leader in medical data standardization in Japan, we will continue to collaborate with various industry players and related organizations, striving to promote the utilization of medical data under the philosophy of "delivering necessary medical care to necessary patients through data."

Comment from NEC In this project, we converted synthetic data of claims, DPC, and HL7 FHIR prescription data into the international common standard OMOP CDM (Common Data Model) and examined the necessary scheme and data utility for the data analysis process actually performed by researchers. Yuimedi provided deep expertise and practical consulting for OMOP conversion, smoothly proceeding with everything from organizing conversion specifications to ensuring quality, defining cohorts based on research questions, and visualizing features, all directly linked to research. We expect that data integration via OMOP will make data easier for researchers to handle and that the establishment of a highly accurate, efficient, and reproducible research environment utilizing standard analysis tools will further promote research activities.

Michio Kaji, Senior Professional, Public Integration Division, NEC Corporation

--- ■ About OMOP CDM OMOP CDM (Observational Medical Outcomes Partnership Common Data Model) is a common data model published by OHDSI (Observational Health Data Sciences and Informatics). It is a data format designed for RWD analysis, characterized by a standardized vocabulary system that allows for the integrated handling of medical terms worldwide. These features enable the unification of RWD from various countries into a common data representation. Furthermore, its normalization for easy storage in databases is another feature suitable for observational research.

--- ■ Company Profile Company Name: Yuimedi Inc. Business Activities: Medical data standardization, construction of medical data utilization networks Representative Director: Emiri Grimes Founded: November 2020 URL: https://yuimedi.com/omop

Contact for this matter Inquiry form URL: https://yuimedi.com/contact