Elderly monitoring systems are gaining significant attention in our increasingly aging society. Existing monitoring systems, which utilize RGB and infrared cameras, often encounter errors when recognizing human-like objects, photos, and videos as actual humans. Additionally, privacy concerns arise due to this issue. However, these challenges can potentially be overcome by employing thermal images. Thus, our study aimed to investigate the feasibility of identifying and categorizing human postures depicted in thermal images using deep learning models and algorithms. To conduct our experiment, we developed a system that utilizes a thermal pose algorithm and a convolutional neural network. As a result, we achieved an average accuracy of 88.3%, with the highest accuracy reaching 91.2%.

The current method of gait analysis has several limitations for determining gait stability, such as a complicated preparation process, repeated experimental procedures that are time-consuming, and financial burden of experiments. This study investigated whether gait stability could be analyzed using only the COM-COP (Center of Mass-Center of Pressure) inclination angle connecting COM and COP. COM and COP coordinates were obtained from a motion analysis system for a total of 40 elderly and young subjects. The COM-COP inclination angle that changed in real time during level walking was then analyzed to obtain gait stability on each of sagittal and frontal planes using these coordinates. As a result, the gait symmetry index on the sagittal plane did not show a statistically significant difference between young and elderly subjects (First Step, p = 0.189; Second Step, p = 0.711). On the frontal plane, elderly subjects showed 0.39 degrees (p = 0.058) and 0.5 degree (p = 0.03) larger side-to-side sway angles in the first and second steps than young subjects, respectively. Gait stability can be analyzed using a more simplified experimental method with minimum amount of data in future gait analysis.

The purpose of this study was to analyze dynamic postural balance against tilting perturbation in the young and the elderly. Twenty-eight young subjects and 22 elderly subjects participated in this study. Subjects performed dynamic balance test on a force plate during tilting perturbations (tilt-up and tilt-down). As outcome measures, peak distance and velocity were calculated from center of pressure (COP). Two-way ANOVA were performed for the outcome measures with the independent factors of age and gender. COP peak distance of the elderly was significantly greater than that of the young (p < 0.05). Velocity of COP showed age difference (p < 0.001) and also interaction effects only in tilt-up perturbation (p < 0.05). Especially, age-related difference existed in only women (p < 0.001). The age-related changing of women in the dynamic balance may be related to the greater fall rate of elderly women.