Wearable sensors are being applied to real-world motion monitoring and the focus of this work is assessing health status and wellbeing. An extensive literature has documented the effects on gait control of impaired physical health, but in this project, the aim was to determine whether emotional states associated with older people’s mental health are also associated with walking mechanics. If confirmed, wearable sensors could be used to monitor affective responses. Lower limb gait mechanics of 126 healthy individuals (mean age 66.2 ± 8.38 years) were recorded using a high-speed 3D motion sensing system and they also completed a 12-item mental health status questionnaire (GHQ-12). Mean step width and minimum foot-ground clearance (MFC), indicative of tripping risk, were moderately correlated with GHQ-12. Ageing and variability (SD) of gait parameters were not significantly correlated with GHQ-12. GHQ-12 scores were, however, highly correlated with left-right gait control, indicating that greater gait symmetry was associated with better mental health. Maintaining good mental health with ageing may promote safer gait and wearable sensor technologies could be applied to gait asymmetry monitoring, possibly using a single inertial measurement unit attached to each shoe.

Globally, we are facing the tendency of aging, and demands for health enhancement among the older population have been steadily increasing. Among various exercise interventions, Pilates has been popularly utilized in rehabilitation; therefore, it is considered suitable for vulnerable populations. In this study, frail late-stage older adults (>75 years) participated in a modified Pilates program (30 min per session, once a week for eight weeks). Age- and condition-matched Controls were also involved as the benchmark to reveal the effect of Pilates. While only the Pilates group participated in the exercise intervention, both groups undertook the health assessments twice (before and after the intervention period). Assessments included: (i) falling risk based on 3D motion capture systems and (ii) overall cognitive functions utilizing Mini-Mental State Examination and executive function with the use of Trail Making Test-A (TMT-A). Two-dimensional mood state was also used to measure changes in mood due to Pilates intervention. An 8-week Pilates intervention was effective in achieving higher and symmetrical swing foot control. Dynamic balance at heel contact was also improved by extending the spatial margin in case of slipping. Despite the trend of positive Pilates effects on executive functions (29% improvement) confirmed by TMT-A, no significant effects were observed for cognitive functions. Positive mood changes were achieved by Pilates intervention, which may be the key for late-stage seniors to continue their participation in exercise programs. While further studies with a larger sample size are essential, Pilates appears to provide adequate exercise for the frail late-stage older population to minimize frailty.

Tripping is the largest cause of falls, and low swing foot ground clearance during the mid-swing phase, particularly at the critical gait event known as Minimum Foot Clearance (MFC), is the major risk factor for tripping-related falls. Intervention strategies to increase MFC height can be effective if applied in real-time based on feed-forward prediction. The current study investigated the capability of machine learning models to classify the MFC into various categories using toe-off kinematics data. Specifically, three MFC sub-categories (less than 1.5 cm, between 1.5 and 2.0 cm, and higher than 2.0 cm) were predicted to apply machine learning approaches. A total of 18,490 swing phase gait cycles’ data were extracted from six healthy young adults, each walking for 5 min at a constant speed of 4 km/h on a motorized treadmill. K-Nearest Neighbor (KNN), Random Forest, and XGBoost were utilized for prediction based on the data from toe-off for five consecutive frames (0.025 s duration). Foot kinematics data were obtained from an inertial measurement unit attached to the mid-foot, recording tri-axial linear accelerations and angular velocities of the local coordinate. KNN, Random Forest, and XGBoost achieved 84%, 86%, and 75% accuracy, respectively, in classifying MFC into the three sub-categories with run times of 0.39 s, 13.98 s, and 170.98 s, respectively. The KNN-based model was found to be more effective if incorporated into an active exoskeleton as the intelligent system to control MFC based on the preceding gait event, i.e., toe-off, due to its quicker computation time. The machine learning-based prediction model shows promise for the prediction of critical MFC data, indicating higher tripping risk.

Tripping is the largest cause of falls and low swing foot ground clearance during the mid-swing phase, particularly at the critical gait event known as Minimum Foot Clearance (MFC) is the major risk factor for tripping-related falls. Intervention strategies to increase MFC height can be effective if applied in real-time based on feed-forward prediction. The current study investigated the capability of machine learning models to classify the MFC into various categories using toe-off kinematics data. Specifically, three MFC sub-categories (less than 1.5cm, between 1.5-2.0cm and higher than 2.0cm) were predicted applying machine learning approaches. A total of 18,490 swing phase gait cycles’ data were extracted from six healthy young adults, each walking for 5-minutes at a constant speed of 4km/h on a motorised treadmill. Both K-Nearest Neighbour (KNN) and Random-Forest were utilised for prediction based on the data from toe-off for five consecutive frames (0.025s duration). Foot kinematics data were obtained from inertial measurement unit attached to the mid-foot, recording tri-axial linear accelerations and angular velocities of the local coordinate. KNN and Random-Forest achieved 84% and 86% accuracy, respectively, in classifying MFC into the three sub-categories with run time of 0.39 seconds and 13.98 seconds respectively. The KNN-based model was found to be more effective if incorporated into an active exoskeleton as the intelligent system to control MFC based on the preceding gait event, toe-off due to its quicker computation time. The machine learning based prediction model shows promise for the prediction of critical MFC data indicating higher tripping risk.