Questions about HDAS

Q: Why is the Biostatistics Division of the Institute of Epidemiology and Preventive Medicine changing to the Institute of Health Data Analytics and Statistics?

A: Given the rapid evolution of the times, both the pace and volume of accumulated health data have far exceeded those of the past. This is particularly evident in the establishment of human biobank research initiatives in various countries. The challenge of how to effectively address this vast accumulation of health data has become an urgent matter to resolve.

For example, in the field of genetics, over the past two decades, we have transitioned from being able to detect only a few genes in a single individual to using microarrays to detect hundreds or even thousands of genes. In recent years, the most significant advancement is the development of the next-generation sequencing (NGS) technology, which can simultaneously analyze tens of thousands of genes within a very short time. With the substantial increase in the number of target genes for analysis, the challenges go beyond the issue of increased false positives due to higher testing frequencies. Many traditional methods are unable to analyze all variables within an acceptable timeframe. Therefore, the development of new analytical and integration algorithms for novel gene data has become a critical objective.

In addition to genetics, the significant advancements in communication technology have enabled the biomedical field to acquire large volumes of long-term individual tracking data through wearable devices, a data collection mode that was previously difficult to achieve in traditional research settings. However, efficiently and accurately analyzing such novel types of health data presents a complex challenge.

Computer technology has also brought about qualitative changes in contemporary health data research. Faced with the accumulation of vast amounts of health data and research reports, we can not only identify commonalities between different datasets through integrated analysis but also explore various causal inference methods to discover essential factors related to health issues.

Lastly, as the healthcare system experiences a significant increase in imaging, speech, and even unstructured text data, the introduction of machine learning methods and artificial intelligence becomes a crucial approach to enhancing analytical efficiency. By integrating these diverse data sources, we believe we can make better contributions to early disease screening and predicting patient survival.

In conclusion, in this era of substantial changes in both the types and quantities of health data, we believe that the transformation from the Biostatistics Division of the Epidemiology and Preventive Medicine Research Institute to the Health Data Analytics and Statistics Research Institute will enable more effective resource allocation and provide superior analytical methods for addressing health data challenges.


Q: What are the differences between NTUHDAS and other statistical research institutes in Taiwan/National Taiwan University?

A: In Taiwan and within the National Taiwan University system, there are several excellent institutes and programs related to statistics. However, currently, the Institute of Health Data Analytics and Statistics (HDAS) is the only institute primarily focused on health data. Located within the National Taiwan University Medical Campus, we have an inherent advantage due to the presence of institutions such as National Taiwan University Hospital, College of Public Health, College of Medicine, College of Pharmacy, and College of Nursing, among others. Collaboration and exchange of resources between these different entities are frequent, and the campus offers not only specialized statistical courses but also a wealth of knowledge in the fields of biomedicine and health data.

We believe that growing in such an environment equips students not only with statistical expertise but also with specialized knowledge in biomedicine and health-related fields. In the era of data science, possessing both statistical analysis skills and domain knowledge is crucial for becoming interdisciplinary professionals in health data analysis. In summary, while maintaining a strong foundation in statistical theory and mathematical modeling, our institute places a greater emphasis on practical applications of health data compared to other research institutes. We anticipate achieving a more balanced approach between theory and practice through collaboration with major healthcare systems and industries, allowing students to directly engage with real-world issues and enhancing their employability and competitiveness in the job market.


Q: What are the differences between NTUHDAS, the Division of Biostatistics of the Institute of Epidemiology and Preventive Medicine?

A: In terms of course offerings, we have a greater emphasis on courses related to mathematics and statistics. In the future, within HDAS, we aim to incorporate more contents related to the domain knowledge from health data. In addition to statistical methods and model development, practical applications of health data will also be a key focus in our research and teaching.


Q: What are the future development directions?

A: HDAS is built upon the basis from the Division of Biostatistics of the Institute of Epidemiology and Preventive Medicine. We will continue to uphold the excellent traditions, focusing on the development of statistical methodology and mathematical models.

In response to the contemporary trends of artificial intelligence and big data, we will collaborate with academia and industry to gradually introduce fields and courses that were previously less emphasized. These may include image analysis, speech recognition, text analysis, and etc. We aim to use real-world challenges in public health and biomedicine as catalysts to drive the development of theory and methodology. Our goal is to ensure that our research outcomes are better aligned with the pressing issues in the current health domain.


Q: Who are the suitable candidates for HDAS?

A: HDAS welcomes students from diverse backgrounds, as health data analysis requires a foundation in various interdisciplinary knowledge areas. For students with backgrounds in statistics and information technology, we provide additional knowledge and practical insights into health data to help them understand the application scenarios of statistical theory and methodology. For students with backgrounds in biomedicine or public health, we offer a stronger foundation in data analysis principles and programming skills, enabling them to conduct hands-on analysis and enhance research design through a deeper understanding of health data.