Partner and Collaborations

Taipei Medical University Clinical Research Database (TMUCRD)

1. Introduction

The Observational Health Data Sciences and Informatics (OHDSI) is a global and non-profit interdisciplinary and open science network headed by Columbia University in the United States. Its vision is to use big data analysis and artificial intelligence methods to enhance the value of clinical medical data, realize cross-domain multi-party research cooperation, and solve problems in the medical field. Currently, participants of the OHDSI global collaboration network include schools (such as Harvard Medical School and Stanford University School of Medicine), hospitals and enterprises (such as IQVIA, Johnson & Johnson, Novartis, Oracle and IBM) from dozens of countries and regions such as the United States, Canada, the United Kingdom and Australia. Currently, the OHDSI has more than 600 million clinical data and has published hundreds of academic papers.

2. History of OHDSI in Asia and Taiwan

OHDSI’s Asia-Pacific Community (APAC) currently includes six regional chapters: South Korea (joined in 2014), China (joined in 2016), Japan (joined in 2019), Australia (joined in 2019), Singapore (joined in 2020) and Taiwan (joined in 2020). The Taipei Medical University Office of Data Science is the first organization in Taiwan to promote OHDSI. After half a year of medical databases integration project, the OHDSI Taiwan Chapter was officially established on December 15, 2020, leading Taiwan in joining the OHDSI alliance. The first OHDSI Asia-Pacific Symposium (APAC Symposium) was held on December 5-6, 2020. It allowed the Asia-Pacific partners (and other partners around the world) to share the latest information and research results in the region, and opened up opportunities for regional events and online discussions. To promote more collaborative partnerships between transnational countries in the Asia-Pacific region, we hold an APAC Community call every two weeks. In addition to providing updated information on regional developments, we also share the latest research reports and conduct collaborative discussions on common interests.

3. Taipei Medical University Clinical Research Database (TMUCRD) international promotion

Since 2015, the Taipei Medical University Office of Data Science has been integrating the electronic medical records databases of TMU’s three affiliated hospitals (Taipei Medical University, Wanfang Hospital and Shuang Ho Hospital) to form the Taipei Medical University Clinical Research Database (TMUCRD). It combines various electronic medical records data of the three hospitals, including structured data (such as patient’s basic information, medical information, test reports, diagnosis results, treatment procedures, surgery and medication status) and unstructured data (such as physicians records, pathology reports, medical imaging reports), and compiled them into analyzable data. The data period covered by TMUCRD is from 1998 to 2020. Since Shuang Ho Hospital joined the TMU System in 2008, the scope of the database includes the complete data of the three hospitals. The data content includes 10 categories , 60 data tables and 2355 fields. In addition, the various data tables can be mutually linked. As of 2019, the database has accumulated the medical information of nearly 3.65 million patients across Taiwan. The Taipei Medical University Office of Data Science has used many opportunities in the past to introduce the TMUCRD to OHDSI countries around the world, such as through oral and poster presentations during online international OHDSI APAC Symposium and APAC Community calls.

4. Use ETL tools to convert clinical database into the international standards OMOP Common Data Model to complete the international integration of the data platform.

The OMOP Common Data Model (CDM) is an internationally unified data standard that regulates the format and content of observational data and supports observational data from different sources. Through the data Extraction-Transformation-Loading (ETL) process, a standardized data structure is formed so that applications such as data query and analysis can be performed accordingly. CDM is designed to meet the scientific research needs of screening and evaluating the relationship between medical interventions (drugs, surgery, policy changes) and curative effects (appearance of a certain condition, surgery, use of a certain drug). Clinical events (diagnosis, observation, surgery ) are required to define certain patient cohorts for treatment or curative effect.


The Taipei Medical University Office of Data Science initiated the OHDSI-CDM grafting project in September 2020. With the assistance of OHDSI headquarters, a series of TMU-OHDSI OMOP CDM transnational grafting technical guidance online meetings were conducted, and in December 2020, the TMUCRD was successfully integrated with the OHDSI-CDM big data platform.

5. Actively participate in numerous large-scale international collaborative research projects

With the construction of the platform for linking the TMUCRD and the OHDSI-OMOP-CDM big data link platform completed, TMU’s clinical data can be internationally integrated with standardized research from 19 countries (more than 100 databases). To date, we have participated in three large-scale multinational collaborative research projects, including a prognostic analysis of continuation therapy with antihypertensive agents, the effectiveness of anticoagulants, and the cancer safety of H2-receptor antagonists. We hope to have more future exchanges and collaborations with international scholars and experts on large-scale international research projects. 

6. Future Developments

2020 is the lead-in period of international cooperation in which TMU linked with the OHDSI alliance through the TMUCRD to jointly solve international clinical practice problems. We have set 2021 as the international collaboration growth stage for TMUCRD. Through participation in a number of international large-scale research projects, we will conduct multi-party collaboration and exchanges with the clinical databases of other advanced countries around the world, thereby realizing the value of TMUCRD. We hope that 2022 will be a period of maturation for TMU’s international collaboration in clinical research. At that time, we plan to lead more international professional partners with our existing research results to conduct transnational data collection and collaboration.

%d bloggers like this: