The effect of camera location on observation-based posture estimation

Evaluation of ergonomic risks for work-related musculoskeletal disorders of nursing tasks in Korea

Among all industrial accident-induced diseases, musculoskeletal disorders (MSDs) are most prominently observed among nurses. The physical load of everyday tasks involved in nursing work was assessed in this study using a developed risk index, whereby the physical burden was evaluated using the exposure duration and work intensity levels. This survey targeted nine small, medium and large-sized hospitals in South Korea and categorized representative nursing tasks into six groups. The subtasks within these six categories (evaluated as high risk) included changing a patient's posture and assisting with walkers or wheelchairs, transporting/handling drug carts, bathing patients, replacing bedding, traction therapy, cardiopulmonary resuscitation and artificial manual breathing unit, and computer work. The risk index ratio was significantly different by task type for each task. We demonstrated that the risk index developed in this study can be used to evaluate MSDs in hospitals.

Measurement of Shoulder Abduction Angle with Posture Estimation Artificial Intelligence Model

Substantial advancements in markerless motion capture accuracy exist, but discrepancies persist when measuring joint angles compared to those taken with a goniometer. This study integrates machine learning techniques with markerless motion capture, with an aim to enhance this accuracy. Two artificial intelligence-based libraries-MediaPipe and LightGBM-were employed in executing markerless motion capture and shoulder abduction angle estimation. The motion of ten healthy volunteers was captured using smartphone cameras with right shoulder abduction angles ranging from 10° to 160°. The cameras were set diagonally at 45°, 30°, 15°, 0°, -15°, or -30° relative to the participant situated at a distance of 3 m. To estimate the abduction angle, machine learning models were developed considering the angle data from the goniometer as the ground truth. The model performance was evaluated using the coefficient of determination R2 and mean absolute percentage error, which were 0.988 and 1.539%, respectively, for the trained model. This approach could estimate the shoulder abduction angle, even if the camera was positioned diagonally with respect to the object. Thus, the proposed models can be utilized for the real-time estimation of shoulder motion during rehabilitation or sports motion.

Ergonomic risk and physiological assessment of plogging activity

BACKGROUND: Plogging, an environment friendly trash workout is a combination of jogging with litter collection. People who are involved in the plogging carry a baggage for collecting the litter. Walking with a weight on one side causes the opposite side of the body to engage for stability and are also exposed to repetitive bending during the activity. OBJECTIVES: The purpose of this study is to evaluate the postural and physiological aspects of plogging activity. METHODS: Thirty six subjects performed the litter collection in stoop, semi-squat, full squat and lunge postures respectively. Postures were analyzed using Rapid Entire Body Assessment (REBA). Physiological aspects of plogging, as well as a comparison of physical activity assessment during jogging and plogging, were investigated using a Polar M430 optical heart rate monitor. Statistical analysis were performed using SPSS version 23. RESULTS: Mean±SD of full squat (5.13±0.59) and lunge (6.64±1.15) posture was found to have lesser risk score in comparison with the other two postures such as stoop (10.31±0.88) and semi-squat (8.11±1.40). Analysis from the Kruskal-Wallis and post hoc test showed that there is no significant interaction between the postures (p < 0.05). Paired Sample t-test showed that the energy expenditure for plogging and jogging are found to be similar (p > 0.05), but the fat percentages of calories burned is more in plogging (p < 0.05). Howerver plogging can be considered as a strenous activity as the % Cardiovascular strain of the activity had a mean value of (99.261%). CONCLUSIONS: Ergonomic interventions are needed to play a vital role in minimizing the musculoskeletal related injuries and the physical strain of the task.

A Mobile Application to Measure Trunk Flexion Angles in Lifting Tasks

Occupational Abstract The aim of this project was to develop and test an application capable of quickly and repeatedly measuring trunk flexion angles during sagittal plane lifting tasks. The developed application uses the built-in accelerometer in mobile devices to approximate trunk flexion angle, as the user follows an operator as they perform a lift. A black line is superimposed over the camera feed, allowing the user to approximate the angle of inclination of a line connecting the operator's seventh cervical and first sacral vertebrae-thereby estimating the trunk flexion angle. The magnitude of this angle and its velocity have been linked to the development of occupational low back pain; thus the application provides ergonomists a more refined means of screening tasks beyond currently available survey tools.

Assessment of the ergonomic risk from saddle and conventional seats in dentistry: A systematic review and meta-analysis

Objective

This study aimed to verify whether the saddle seat provides lower ergonomic risk than conventional seats in dentistry.

Methods

This review followed the PRISMA statement and a protocol was created and registered in PROSPERO (CRD42017074918). Six electronic databases were searched as primary study sources. The "grey literature" was included to prevent selection and publication biases. The risk of bias among the studies included was assessed with the Joanna Briggs Institute Critical Appraisal Tool for Systematic Reviews. Meta-analysis was performed to estimate the effect of seat type on the ergonomic risk score in dentistry. The heterogeneity among studies was assessed using I² statistics.

Results

The search resulted in 3147 records, from which two were considered eligible for this review. Both studies were conducted with a total of 150 second-year dental students who were starting their laboratory activities using phantom heads. Saddle seats were associated with a significantly lower ergonomic risk than conventional seats [right side (mean difference = -3.18; 95% CI = -4.96, -1.40; p < 0.001) and left side (mean difference = -3.12; 95% CI = -4.56, -1.68; p < 0.001)], indicating posture improvement.

Conclusion

The two eligible studies for this review provide moderate evidence that saddle seats provided lower ergonomic risk than conventional seats in the examined population of dental students.

Observer variability in posture assessment from video recordings: The effect of partly visible periods

Observers rank partly visible postures on video frames differently than fully visible postures, but it's not clear if this is due to differences in observer perception. This study investigated the effect of posture visibility on between-observer variability in assessments of trunk and arm posture. Trained observers assessed trunk and arm postures from video recordings of 84 pulp mill shifts using a work sampling approach; postures were also categorized as 'fully' or 'partly' visible. Between-worker, between-day, and between-observer variance components and corresponding confidence intervals were calculated. Although no consistent gradient was seen for the trunk, right upper arm posture showed smaller between-observer variance when all observers rated a posture as fully visible. This suggests that, partly-visible data, especially when observers disagree as to the level of visibility, introduces more between-observer variability when compared to fully visible data. Some previously-identified differences in daily posture summaries may be related to this phenomenon.

Observer performance in estimating upper arm elevation angles under ideal viewing conditions when assisted by posture matching software

Selecting a suitable body posture measurement method requires performance indices of candidate tools. Such data are lacking for observational assessments made at a high degree of resolution. The aim of this study was to determine the performance (bias and between- and within-observer variance) of novice observers estimating upper arm elevation postures assisted by posture matching software to the nearest degree from still images taken under ideal conditions. Estimates were minimally biased from true angles: the mean error across observers was less than 2°. Variance between observers was minimal. Considerable variance within observers, however, underlined the risk of relying on single observations. Observers were more proficient at estimating 0° and 90° postures, and less proficient at 60°. Thus, under ideal visual conditions observers, on average, proved proficient at high resolution posture estimates; further investigation is required to determine how non-optimal image conditions, as would be expected from occupational data, impact proficiency.

Partly visible periods in posture observation from video: prevalence and effect on summary estimates of postures in the job

This paper investigated the extent to which observers rated clearly visible postures on video differently from partly visible postures, and whether visibility affected full-shift posture summaries. Trunk and upper arm postures were observed from 10,413 video frames representing 80 shifts of baggage handling; observers reported postures as fully or only partly visible. Postures were summarized for each shift into several standard metrics using all available data, only fully visible frames, or only partly visible frames. 78% of trunk and 70% of upper arm postural observations were inferred. When based on all data, mean and 90th percentile trunk postures were 1.8° and 5.6° lower, respectively, than when based only on fully visible situations. For the arm; differences in mean and 90th percentile were 0.7° and 8.2°. Daily posture summaries were significantly influenced by whether partly visible postures are included or not.