Analyzing Walking and Driving Behavior Across Different Age Groups Using Population Analysis and Correlation Networks

Altered mobility patterns are one of the early symptoms of aging. Several motor-related disorders such as Parkinson's disease are also associated with decreased or altered mobility. However, there is no standard clinical test to identify decreased mobility or diagnose movement variations. In recent years, wearable devices have become popular in measuring mobility parameters and quantifying physical activities such as walking and driving. Furthermore, wearable devices have been widely used in the collection and analysis of mobility data because they are small, affordable, and easy to use. The main objective of this research is to develop a data-driven computational model that can analyze mobility data collected from a group of individuals from different age groups and capture potential aging-related movement variabilities. Such a model would also allow computational tools to extract meaningful correlations between age groups and physical activity. In this study, we have analyzed the mobility data collected from 32 healthy adults from different age groups while they are walking outdoors, climbing stairs, and driving. We developed correlation network models and employed a population analysis-based approach to unravel the associations between aging and physical activities; mainly walking and driving. Although our analysis produced interesting results and identified key parameters that impacted the walking and driving patterns, it didn't significantly different between subjects belonging to different age groups, which we believe is mainly due to the limited use dataset in terms of size and variability. However, the proposed model and recommended approach pave the way for future studies that will further explore the relationships between mobility and aging using richer datasets.