Work-Related Risk Assessment According to the Revised NIOSH Lifting Equation: A Preliminary Study Using a Wearable...

Area of Science:

• Occupational Health and Safety
• Biomechanics
• Machine Learning Applications

Background:

• Lifting heavy loads is a common cause of work-related musculoskeletal disorders.
• The National Institute for Occupational Safety and Health (NIOSH) developed a quantitative method for assessing lifting risks.
• Accurate biomechanical risk assessment is crucial for effective prevention strategies.

Purpose of the Study:

• To investigate the feasibility of using machine learning (ML) to classify biomechanical risk.
• To evaluate ML algorithms for classifying lifting tasks based on the revised NIOSH lifting equation.
• To develop an automated method for assessing work-related biomechanical exposure.

Main Methods:

• Collected acceleration and angular velocity data using wearable sensors during lifting tasks.
• Extracted time-domain features (RMS, min, max, std dev) from sensor signals.
• Applied and evaluated various ML algorithms for binary risk classification.

Main Results:

• Tree-based ML algorithms achieved over 90% accuracy in risk classification.
• Area under the ROC curve exceeded 0.9 for the best-performing algorithms.
• The proposed method demonstrated high efficacy in distinguishing between risk and no-risk lifting scenarios.

Conclusions:

• The combination of extracted features and ML algorithms provides a valuable approach for automated biomechanical risk classification.
• This methodology shows significant potential for assessing subject exposure to biomechanical risks during work activities.
• The findings support the use of ML for enhancing occupational safety and preventing musculoskeletal injuries.