My name is Shigeyuki Matsui, and I have been appointed to the Department of Biostatistics, Kyoto University School of Public Health since April 1, 2025. I graduated from the Department of Industrial Administration (now the Department of Information and Computer Technology) at the Faculty of Engineering, Tokyo University of Science, where I received my doctoral degree under the guidance of Professors Toshiro Haga and Isao Yoshimura. After working for a pharmaceutical company and serving as a research assistant and full-time lecturer at Oita University of Nursing and Health Sciences, I served as an associate professor at the Department of Pharmacoepidemiology, Kyoto University School of Public Health since 2002 (6 years), associate professor and professor at the Institute of Statistical Mathematics, Research Organization of Information and Systems since 2008 (5 years), and professor at the Department of Biostatistics, Nagoya University Graduate School of Medicine, since 2013 (12 years). And now, after 17 years, I am once again working at Kyoto University. I am feeling very emotional as I recall my life in Kyoto in my late 30s.

Although my background is in engineering, my research field since graduate school has consistently been biostatistics. In particular, I have studied statistical methods and their practices in the design and data analysis of medical research involving human subjects. As a practical science, statistics is not only about statistical theory and mathematical research (corresponding to the basic research in biostatistics), but also about practical research as its application, and the feedback from that leads to new basic research, and then that practical research… It is important to keep going through this cycle. Medical research on human subjects includes evaluation of prevention, diagnosis, and treatment in a wide range of disease areas, so the medical research that is the subject of biostatistics is extremely diverse. It has been more than 30 years since my graduate school days, but I continue to be overwhelmed by the breadth and depth of the problems that biostatistics deals with.

A topic of particular interest in recent years has been the design and data analysis of clinical studies for personalized medicine that combine diagnostic and therapeutic methods. This theme is the intersection of the traditional statistical framework of statistical inference and the recent development of machine learning and predictive analytics. In particular, there are many unique statistical challenges in data-driven diagnostics development and validation under limited biological knowledge. On the other hand, I am working on constructing data science for handling small-scale data, such as improving efficiency of experimental studies with limited research resources through adaptive experimental design, utilizing external information through Bayesian and transfer learning methods, and conditional inference (e.g., selective inference) that takes these factors into account. Other important directions include data-driven causal inference and dynamic therapeutics development in observational studies that deal with large-scale data. As such, I would like to establish a new data science by integrating traditional statistics with machine learning and AI on a wide range of topics.

In terms of human resource development, we have high hopes for the potential of medical students, particularly those from School of Medicine, but to secure stable data science human resources, we need to actively try to recruit students from other faculties such as mathematics, informatics, and economics. I would like to do my utmost to form a strong basis for human resource development that attracts future data science talents with such a wide range of backgrounds to enter the medical and healthcare fields and steadily produces many medical data science specialists.

Of course, most of the above activities would not be possible without the help and support of many people. I look forward to your continued guidance and support.