AI model estimates biological age through 5 drops of blood

In fact, there are some people who seem to challenge time and look younger than people of the same age. The group of scientists at Osaka University, Japan, may have found a way to quantify this difference. By putting hormonal transformers into a AI model, they develop the system that allows biological age estimates – a measure of the body’s aging level – instead of counting age.

Using only 5 drops of blood, the new system analyzes 22 types of main steroid hormones and their interaction to give health more accurate health assessments. New research published in magazines Science Advances March 14.

Aging is not only the number of years per person who has lived but is shaped by genetics, lifestyle and environmental factors. Traditional methods to estimate biological age based on major biological signs, such as protein levels or DNA methylation, but often ignore sophisticated hormone networks that regulate the body’s balance.

“The human body relies on hormones to maintain the lip balance. So we think, why not use them as the main indicator of aging?”, Dr. Qiuyi Wang, co -author of the study, said. To check this idea, the team focused on steroid hormones, playing an essential role in metabolism, immune function and reaction to stress.

The research team developed a deep artificial neural network model (DNN) combining steroid metabolism paths, becoming the first AI model to explain the interaction between different steroid molecules. Instead of considering absolute steroids – significantly changing between individuals – a new model to check the steroid ratio, providing more accurate and personalized assessments of biological age. The model is trained based on blood samples from hundreds of individuals, revealing that the difference in biological age tends to increase as people get older.

One of the most prominent findings of a new study related to cortisol, a steroid hormone is often associated with stress. The team found that when cortisol level doubled, biological age increased by 1.5 times. This shows that chronic tension can accelerate the aging process at the biochemical level, strengthen the importance of stress management to maintain long -term health.

The group of scientists believes that this model of AI biological age can pave the way for health monitoring in a more personalized way. Future applications may include early detection of disease, developing custom health care programs, making separate recommendations about each individual’s lifestyle to slow down the aging process.

Although the new research is a big step, experts believe that biological aging is a complicated process affected by many factors outside the hormone. “This is only the beginning. By expanding the data set and integrating additional biological signs, we hope to improve the model further and gain more profound understanding of the aging mechanism,” said Dr. Zi Wang, co -authored.