id_789. A MACHINE LEARNING-BASED APPROACH TO ASSESSING THE EFFECTS OF BALANCE AND COGNITIVE-MOTOR TRAINING IN REAL AND EXTENDED ENVIRONMENTS IN SENIORS.
Beata Sokołowska1, Teresa Sadura-Sieklucka2, Ewa Sokołowska3
1 Mossakowski Medical Research Institute, Polish Academy of Sciences, 5 Pawińskiego St., Warsaw, Poland
2 National Institute of Geriatrics, Rheumatology and Rehabilitation, 1 Spartańska St., Warsaw, Poland
3 Institute of Psychology, John Paul II Catholic University of Lublin, 14 Aleje Racławickie St., Lublin, Poland
INTRODUCTION: Computational neuroscience not only bridges the gap between theoretical models and experimental studies to validate them, but also offers advanced computational tools (machine learning and artificial intelligence algorithms, ML/AI) for example, to assess beneficial and adverse effects using highly attractive digital environments such as extended reality (virtual, augmented, and mixed reality, XR:VR/AR/MR). Their application in (neuro)geriatrics is particularly promising, given the rapid aging of modern societies and the serious health consequences associated with it.
AIM(S): The main goal of the research was to evaluate the impact of various training on seniors.
METHOD(S): The study included 76 individuals aged 60 and above who participated in conventional exercises using various equipment (control group) or cognitive-motor exercises using the ActivLife virtual platform (AR group). The results of clinical geriatric tests before and after these exercise sessions were used as learning sets in an ML approach based on the standard k-nearest neighbors (k-NN) decision rule.
RESULTS: After completing the exercise program, both groups demonstrated similar, statistically significant improvements in all clinical tests and a reduced risk of falls. Additionally, participants in the AR group showed greater engagement in the training sessions.
CONCLUSIONS: Unlike traditional cognitive-motor exercise methods, XR technologies with ML/AI offer interactive training, including remote training, and provide feedback on progress and corrected mistakes. This innovative multifaceted approach allows seniors to choose cognitive and motor exercises tailored to their abilities and needs, which they can modify or interrupt themselves, or continue exercises in a friendlier environment, such as at home. The use of extended environments and advanced computing tools in exercise/training has enormous potential in promoting healthy aging and preventing cognitive-motor function disturbances, as well as balance loss and falls among older adults.