An analysis of 3D motion recordings from more than 3,000 volunteers has shown that the way the body and torso move while walking is linked to the state of internal organs and the quality of nighttime rest, the website infohub.kz reports.

Researchers at the Weizmann Institute in Israel have developed an artificial intelligence system that can use gait and whole-body movement patterns to detect hidden health indicators, including liver condition, diabetes and sleep disorders. The findings were published in the journal Nature Medicine, according to ScienceAlert.

For the study, a team led by computer scientist Adam Gabet analyzed motion recordings from 3,328 adult volunteers. Participants performed five tasks: walking on a treadmill at 3 kilometers per hour and at their usual pace, marching in place, standing up from a chair, and standing still to test balance. A depth camera recorded the position of 26 joints in three-dimensional space, after which a neural network learned to reconstruct hidden portions of the movements and convert them into numerical models.

Comparing the data with medical records and test results revealed a statistically significant link between gait and 1,887 of the 3,210 health characteristics studied. Movement patterns were found to assess liver tissue stiffness — an indicator of scarring — at least as accurately as blood tests. In addition, upper-body movements reflected the quality of the previous night's sleep, insomnia and the risk of sleep apnea, and improved the accuracy of predicting diabetes and depression.

According to the authors, such a contactless test could in the future complement standard medical checkups, once clinical trials are conducted in more diverse populations.