Artificial Intelligence in Predicting Biomechanical Effects of Striatal - Locomotor Tasks: Application in Fall Risk Assessment and Rehabilitation

Falls are one of the leading causes of severe injuries in older adults, making the identification of fall risk using modern technologies particularly important. Systematic studies aimed at determining the optimal combination of sensor placement and movement features have shown that gait speed, measured using sensors mounted on the shank, is the most effective indicator for assessing fall risk during walking. Additionally, linear acceleration measured at the lower back is crucial for evaluating balance while standing still and during sit-to-stand transitions. Meta-analyses have identified four key features that are more prevalent in individuals at risk of falling: the root mean square (RMS) acceleration in the mediolateral direction, step count, Timed Up and Go (TUG) test completion time, and step duration. Four main sensor technologies have been proposed for fall risk detection: inertial sensors, video cameras, force plates, and laser sensors. However, inconsistencies in assessment tools and modeling approaches have prevented the establishment of a global standard. Various tools have been developed to assess functional balance. The Berg Balance Scale (BBS) and Timed Up and Go (TUG) test are among the most validated methods. Other tests, such as the Functional Reach Test (FRT) and the Four Square Step Test (FSST), also demonstrate high reliability. However, further research is needed to standardize these tests and enhance the accuracy of fall risk prediction.

کلیدواژه ها:

Artificial Intelligence in Healthcare and Rehabilitation ، Machine Learning for Fall Prediction ، Fall Risk Assessment in Elderly Populations ، AI in Biomechanics