A “diabetes risk prediction model” tailored for Koreans has come out: Dong-A Science

UNIST · Research on the Gospel hospital of Kosin University

An overview of the development of a machine learning model to predict the incidence of type 2 diabetes specific to Koreans. Provided by UNIST

A machine learning model (machine learning) has been developed to predict the onset of “type 2 diabetes” specifically for Koreans. It should be used to predict the onset of type 2 diabetes in patients in hospitals.

The Ulsan National Institute of Science and Technology (UNIST) announced on the 5th that a joint research team led by Professor Lee Jung-hye of the Department of Industrial Engineering and Professor Kang Ji-hoon of Kosin University Gospel Hospital has developed a model learning algorithm that improved the predictive performance of type 2 diabetes onset based on a large-scale Korean cohort.

Type 2 diabetes is a disease caused by a combination of genetic and environmental factors. It occurs because the ability of pancreatic beta cells to secrete insulin is limited. It is a common disease that affects 1 in 6 Koreans over the age of 30.

Existing studies on diabetes risk prediction models were mainly aimed at the Western population. Thus, there were limitations in predicting diabetes that reflected genetic and environmental factors specific to Koreans.

The research team used a large-scale cohort from the Korean Genome Epidemiology Survey (KoGES) collected by the National Institutes of Health at the Centers for Disease Control and Prevention to develop a predictive model using information specific to Koreans. This cohort has been collected since 2001 for the study of chronic diseases such as metabolic syndrome.

The research team developed a Korean-specific polygenic risk score (gPRS) across a cohort. Here, demographic information, clinical information, and metabolomic information were used together. The predictive model finally developed for type 2 diabetes showed approximately 11% higher predictive performance than when using demographic information alone. Compared to the case in which demographic information and clinical information were also used, the predictive performance improved by about 4% or more.

The research team explained that using the model developed this time, it is possible to identify the risk and occurrence factors of diabetes specific to Koreans. Type 2 diabetes is expected to be effectively prevented and resolved in clinical settings.

Professor Jeong-Hye Lee, who led this study, said: ‘It is very significant to change the approach from a study focusing on the Western cohort to a Korean cohort.’ The results of the study were published on the 29th of last month in the international academic journal ‘eBioMedicine’.

Courtesy of Getty Images bank

Courtesy of Getty Images bank

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Recent News

Editor's Pick