With the increasing aging population, nursing care providers have been facing a substantial risk of work-related musculoskeletal disorders (WMSDs). Visual-based pose estimation methods, like OpenPose, are commonly used for ergonomic posture risk assessment. However, these methods face difficulty when identifying overlapping and interactive nursing tasks, resulting in missing and misidentified skeletons. To address this, we propose a skeleton compensation method using improved spatial temporal graph convolutional networks (ST-GCN), which integrates kinematic chain and action features to assess skeleton integrity and compensate for it. The results verified the effectiveness of our approach in optimizing skeletal loss and misidentification in nursing care tasks, leading to improved accuracy in calculating both skeleton joint angles and REBA scores. Moreover, comparative analysis against other skeleton compensation methods demonstrated the superior performance of our approach, achieving an 87.34% REBA accuracy score. Collectively, our method might hold promising potential for optimizing the skeleton loss and misidentification in nursing care tasks.

The six-minute walking test (6MWT) is an essential test for evaluating exercise tolerance in many respiratory and cardiovascular diseases. Frailty and sarcopenia can cause rapid aging of the cardiovascular system in elderly people. Early detection and evaluation of frailty and sarcopenia are crucial for determining the treatment method. We aimed to develop a wearable measuring system for the 6MWT and propose a method for identifying frailty and quantifying walking muscle strength (WMS). In this study, 60 elderly participants were asked to wear accelerometers behind their left and right ankles during the 6MWT. The gait data were collected by a computer or smartphone. We proposed a method for analyzing walking performance using the stride length (SL) and step cadence (SC) instead of gait speed directly. Four regions (Range I–IV) were divided by cutoff values of SC = 2.0 [step/s] and SL = 0.6 [m/step] for a quick view of the frail state. There were 62.5% of frail individuals distributed in Range III and 72.4% of non-frail individuals in Range I. A concept of a WMS score was proposed for estimating WMS quantitatively. We found that 62.5% of frail individuals were scored as WMS1 and 41.4% of the non-frail elderly as WMS4. The average walking distances corresponding to WMS1–4 were 207 m, 370 m, 432 m, and 462 m, respectively. The WMS score may be a useful tool for quantitatively estimating sarcopenia or frailty due to reduced cardiopulmonary function.

There is a high risk of musculoskeletal discomfort and injury due to the lack of professional guidance and training in caregiving postures. This study aimed to develop a risk assessment and visualization method by analyzing caregiving postures. Participants with (n = 8) and without (n = 10) caregiving experience were recruited to simulate patient transfer from bed to wheelchair. The Rapid Entire Body Assessment (REBA) method lacked sensitivity in distinguishing the experienced and inexperienced groups. We found that the visualization of the center of gravity (COG) trajectory could represent distinct posture differences between the two groups. Based on this finding, we considered a modified REBA method combining the COG trajectory, load-bearing time, and asymmetric load parameters, named the Caregiving-REBA (C-REBA) method. Our results demonstrated that C-REBA could effectively distinguish experienced and inexperienced caregivers, especially in caregiving task Stages 2–4. In conclusion, the present work explored adjusting to the parameters of the REBA method. The proposed C-REBA method could be easily imbedded into the Internet of Things (IoT) device to assess the caregiving posture for providing visual guidance and warning of the risk of discomfort or injury.