Compensation Method for Missing and Misidentified Skeletons in Nursing Care Action Assessment by Improving Spatial...

Area of Science:

• Ergonomics and Occupational Health
• Computer Vision and Machine Learning
• Healthcare Technology

Background:

• The aging population increases the risk of work-related musculoskeletal disorders (WMSDs) among nursing care providers.
• Current visual-based pose estimation methods (e.g., OpenPose) struggle with overlapping and interactive nursing tasks, leading to skeleton misidentification and data loss.
• Accurate ergonomic posture risk assessment is crucial for preventing WMSDs in healthcare settings.

Purpose of the Study:

• To develop and validate a novel skeleton compensation method for improving the accuracy of ergonomic risk assessment in nursing tasks.
• To address the limitations of existing pose estimation techniques in complex, interactive caregiving scenarios.
• To enhance the reliability of calculating joint angles and risk scores (e.g., REBA) in nursing environments.

Main Methods:

• Proposed an improved spatial temporal graph convolutional network (ST-GCN) for skeleton compensation.
• Integrated kinematic chain and action features to assess and correct skeleton integrity.
• Evaluated the method's performance in optimizing skeletal loss and misidentification during nursing tasks.

Main Results:

• The proposed ST-GCN method effectively optimized skeletal loss and misidentification in nursing care tasks.
• Demonstrated improved accuracy in calculating skeleton joint angles and REBA scores.
• Achieved a superior REBA accuracy score of 87.34% compared to other skeleton compensation methods.

Conclusions:

• The developed skeleton compensation method shows significant potential for enhancing the accuracy of ergonomic assessments in nursing.
• This approach offers a promising solution for mitigating WMSDs by improving the reliability of pose estimation in complex healthcare tasks.
• Further application of this method could lead to optimized safety protocols and reduced occupational risks for nursing staff.