Mobility assessments and machine learning aid aging analysis

Research and clinical practices are increasingly focusing on mobility assessments to monitor aging and prevent falls in older adults.

A study utilizing machine learning and inertial sensors compared the gait patterns of young adults and older adults. The research found that aging in functionally healthy individuals manifests primarily through temporal dimensions of walking, such as stance, swing, and stride times, rather than spatial parameters. The Random Forest model proved most effective in classifying these differences.

In clinical settings, mobility tests like the “sit-to-stand” and the “Timed Up and Go” (TUG) are used to evaluate strength, balance, and resistance. Experts note that while these tests are not standalone diagnostics, they serve as vital indicators for healthcare professionals to investigate underlying causes of functional decline, such as cardiovascular issues, fatigue, or loss of balance.