prediction, how OHDSI impacts global collaboration, the community growth in the APAC region, and plenty more.

Congratulations on recently earning your PhD at Taipei Medical University. Can you share an overview of your research focus and key findings?

My research focuses on predicting long-term complications on patients with type 2 diabetes including cardiovascular diseases and cognitive impairment. Using the Taipei Medical University Clinical Research Database and AI-driven models, I’ve developed tools to identify individuals with type 2 diabetes who are at higher risk of developing dementia, which is published on JMIR. A key finding from my work is that electronic health records (EHRs)—including demographics, comorbidities, medications, and lab results—significantly improve prediction accuracy. This highlights the power of data-driven approaches in addressing the growing challenges of neurodegenerative diseases in aging societies.

How did you first get involved with the OHDSI community, and how has it influenced your research?

I discovered OHDSI during my PhD at TMU under the mentorship of Dr. Jason C. Hsu, head of OHDSI Taiwan. I was drawn to its commitment to open science and the OMOP Common Data Model (CDM), which provides a standardized way to harmonize and analyze healthcare data. Being part of OHDSI has transformed my research, giving me tools to develop robust, reproducible patient-level prediction models and collaborate on large-scale studies. Beyond the technical benefits, OHDSI’s collaborative spirit has connected me with brilliant researchers worldwide, helping me refine my methods and broaden my perspective.

You led dementia prediction research that was recently published in the Journal of Medical Internet Research. What makes OHDSI’s approach to patient-level prediction models so effective in generating reliable evidence?

OHDSI’s strength lies in its focus on standardized, reproducible research. The OMOP CDM ensures data from different sources can be analyzed consistently, enabling models to be validated across populations. Its patient-level prediction framework incorporates best practices in machine learning, such as automated feature selection, transparent evaluation, and open-source tools, ensuring accurate and generalizable predictions.

OHDSI has been expanding rapidly in the Asia-Pacific region. What excites you most about this growing collaboration, and what recent developments can you share?

The rapid growth of OHDSI in Asia-Pacific is particularly exciting. Countries like Taiwan, South Korea, Singapore, and Australia are adopting the OMOP CDM and strengthening data-sharing efforts, paving the way for deeper insights into disease patterns and treatments across the region. Vietnam is also planning its first OMOP conversion, which will enable participation in global OHDSI studies. Regional chapters play a vital role in adapting OHDSI’s methods to local healthcare contexts, fostering collaboration, and advancing open science in diverse settings.

You have been involved in several network studies over the years. What does OHDSI do particularly well in fostering global collaboration and generating real-world evidence?

OHDSI’s collaborative, open-science environment empowers researchers to conduct large-scale studies while maintaining patient privacy. Tools like ATLAS and HADES make research transparent and reproducible, combining expertise from various disciplines to produce rigorous and clinically meaningful real-world evidence.

What are some of your hobbies, and what is one interesting thing that most community members might not know about you?

Outside of my professional work, I enjoy outdoor activities like running, cycling, and hiking. One of my favorite memories is when all the OHDSI collaborators gathered for drinks after a major event. It was such a wonderful time, chatting and connecting as friends in a relaxed and fun atmosphere.