BEIJING-Chinese researchers are using artificial intelligence (AI) to classify different types of diabetes, which may help patients obtain more precise treatment.

Different types of diabetes require diverse treatment. The current diabetes classification system, which has been used for more than 20 years, is based on cause and pathological features, which has limitations in guiding clinical treatment.

Researchers from Peking University People's Hospital are working on a more elaborate classification of diabetes that may support individualized treatment. They conducted research on diabetes classification based on the data of 2,316 Chinese people newly diagnosed with diabetes and 815 Americans.  Using the AI clustering method, they separate the two groups into four diabetes subtypes based on five variables including age, BMI, blood glucose levels and insulin resistance indexes.

According to Zou Xiantong, one of the researchers, a previous study from Northern Europe has used similar methods to divide diabetes into five subgroups and demonstrated that the subgroups have different clinical manifestations and corresponding treatments. However, all cases involved in the study were from Northern Europe, and it is unknown whether it is applicable to other populations.

  "We hope our research may provide data support for more accurate typing and treatment of diabetes in the Chinese population," Zou said.

 The data analysis showed that the main clinical features of the four subtypes were basically consistent in the Chinese and U.S. groups, which also coincided with the subtype characteristics of the Northern Europe research. The research was published in the journal The Lancet Diabetes and Endocrinology.