Accuracy of Automatically Identifying the American Conference of Governmental Industrial Hygienists Threshold Limit Values Twelve Lifting Zones over Three Simplified Zones Using Computer Algorithm

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 1; p. 111
• Main Authors: Barim, Menekse S., Lu, Ming-Lun, Feng, Shuo, Hayden, Marie A., Werren, Dwight
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 01.01.2025; MDPI
• Subjects: Accuracy; ACGIH TLVS for lifting; Algorithms; Biomechanics; computer algorithm; Conferences, meetings and seminars; IMU; Laboratories; lifting; Machine learning; Motion capture; Musculoskeletal diseases; Risk factors; Sensors; Variables; United States; Ohio; United States > US; IMU; ACGIH TLVS for lifting; lifting; computer algorithm
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s25010111

The American Conference of Governmental Industrial Hygienists (ACGIH) Threshold Limit Values (TLVs) for lifting provides risk zones for assessing two-handed lifting tasks. This paper describes two computational models for identifying the lifting risk zones using gyroscope information from five inertial measurement units (IMUs) attached to the lifter. Two models were developed: (1) the ratio model using body segment length ratios of the forearm, upper arm, trunk, thigh, and calf segments, and (2) the ratio + length model using actual measurements of the body segments in the ratio model. The models were evaluated using data from 360 lifting trials performed by 10 subjects (5 males and 5 females) with an average age of 51.50 (±9.83) years. The accuracy of the two models was compared against data collected by a laboratory-based motion capture system as a function of 12 ACGIH lifting risk zones and 3 grouped risk zones (low, medium, and high). Results showed that only the ratio + length model provides acceptable estimates of lifting risk with an average of 69% accuracy level for predicting one of the 3 grouped zones and a higher rate of 92% for predicting the high lifting zone.