Scientists have created a clock that predicts diseases years before they begin

Researchers from Beijing University presented the results of the LifeClock project - an artificial intelligence-based model capable of not only determining a person's biological age but also predicting possible diseases and life expectancy. Medical data from 9.6 million people were used to train this system. The algorithm took into account 184 parameters recorded at different life stages - from birth to death. These included anthropometric indicators, medical history, and results of laboratory and instrumental studies.

As BAKU.WS reports with reference to the foreign press, the analysis showed that biological age can be conditionally divided into two stages: development (up to approximately 18 years) and aging. The biomarkers affecting these processes differ. For instance, the "development clock" is most strongly influenced by growth indicators, creatinine levels, and total protein, while the "aging clock" is affected by the concentration of urea, albumin, and the RDW indicator (red blood cell distribution width, reflecting the proportion of blood cells with size deviations).

During experiments, the LifeClock model successfully predicted growth delays, developmental disorders, obesity, pituitary hypofunction, and other conditions in children. In adults, the system accurately forecasted the risk of developing type 2 diabetes, stroke, kidney failure, and cardiovascular diseases.

According to scientists, this technology will allow doctors to identify at-risk patients in advance and conduct preventive measures even before symptoms appear. This is especially important for middle-aged people. Chronological age, determined by date of birth, does not reflect the degree of body wear and accumulated damage, which is why biological age is considered a more accurate indicator of health status.

Previously existing methods for assessing biological age required complex and expensive studies, including genetic tests and molecular analysis. However, such approaches only provide information about the current state of the organism and do not take into account the dynamics of changes over time, meaning they cannot predict diseases - unlike the LifeClock model.