St. Peter's Hospital Director Lee, Choon Sung publishes 'Getting Started with Bayesian Statistics'

 St. Peter's Hospital Director Lee, Choon Sung publishes 'Getting Started with Bayesian Statistics'
Bayesian statistics basics explained for non-specialists
Introduces uses in AI, machine learning and medical decision-making
Explores where clinical uncertainty meets statistical thinking
Reporter Choi In-hwan Posted 2026.06.12 09:44


Copy
 Subscribe
Courtesy of St. Peter's Hospital
Courtesy of St. Peter's Hospital
Lee, Choon Sung, Director of the Spine Center at St. Peter's Hospital, has published 'Getting Started with Bayesian Statistics', an introductory book on the basic concepts of Bayesian statistics. The book explains, in terms a non-specialist can follow, the concepts of Bayesian statistics — which treats uncertain judgment mathematically — and how it can be applied in clinical practice.

St. Peter's Hospital announced on the 12th that Lee, Choon Sung, Director of its Spine Center, has published an introduction to Bayesian statistics, ‘Getting Started with Bayesian Statistics’.

Bayesian statistics is a method of statistical analysis that corrects existing judgments and improves predictive accuracy as data accumulate. Because statistical inference can be updated continuously to reflect new information, it is used to tackle real-world decision-making problems with a high degree of uncertainty.

Recently, Bayesian statistics has also been used in data-driven advanced technology fields such as AI and machine learning. In clinical practice, too, it is increasingly applied across areas such as clinical trial design, prediction of treatment risk for individual patients, estimation of diagnostic accuracy and prevalence, and analysis of drug side effects.

Director Lee first encountered Bayesian statistics ten years ago while at Asan Medical Center and has pursued research on it since, drawn to where it meets clinical judgment.

In clinical practice, rather than judging a patient from a single test, physicians repeatedly refine the diagnosis and the direction of treatment by combining symptoms, medical history, risk factors and response to treatment. Director Lee saw that this process has a structure similar to Bayesian statistical thinking, and was drawn to the fact that clinical judgment can be described in the language of probability.

The book is an introduction that sets out the basic concepts clearly so that even readers without a background in statistics can understand Bayesian statistics. Drawing on the difficulties the author himself experienced as a non-specialist in statistics, it is structured so that readers new to Bayesian statistics can approach it with ease.

The book covers the history of Bayesian statistics, Bayesian statistical thinking and how it is applied in clinical medicine. The related material is treated broadly, the publisher explains, so that readers can explore the concepts together with their practical use.

Frequentist statistics is still more widely used in medical research and clinical practice. The p-value, used to judge the statistical significance of results, is a prime example.

However, as super-aging brings more elderly patients and patients with multiple conditions, and as diseases appear at younger ages, more flexible and comprehensive medical judgment is becoming increasingly important. Director Lee explained that amid such changes in the healthcare environment, Bayesian statistical thinking can serve as an approach to systematic clinical judgment.

Director Lee, Choon Sung said, "Bayesian statistics is a tool that helps you reach better decisions and judgments even under uncertain conditions and with limited information, and it can be usefully applied to solving a wide range of real-world problems, including in clinical practice."

He went on, "Through this book I hope many more people will come to understand the basic concepts of Bayesian statistics and encounter a new perspective on statistical thinking."