Risk Assessment of Musculoskeletal Disorders Using Artificial Intelligence

Abstract

Agricultural ergonomics employs methods such as Rapid Upper Limb Assessment (RULA) and Rapid Entire Body Assessment (REBA) to assess postural risks. However, these methods may be inaccurate and time-consuming. The objective of this study is to compare the effectiveness of Artificial Intelligence (AI), specifically a software based on MediaPipe, with conventional methods (RULA-REBA) to identify and assess ergonomic risks due to postures in rice agriculture. The methodology employed involved the development of AI software with MediaPipe, which was designed to detect postures in real time. This model was capable of identifying 33 anatomical points, thereby enabling detailed analysis of movement and posture. The results demonstrated that the AI outperformed RULA and REBA in detecting forced postures. Furthermore, it provided faster and more accurate assessments. The findings indicated that AI could be a valuable tool in agricultural ergonomics, potentially outperforming traditional methods. This could significantly improve working conditions and reduce musculoskeletal disorders among farmers.