Development of a real-time kinetic-kinematic model to determine the load in the lumbosacral joint

Abstract

Introduction: Unergonomic postures and movements combined with possibly large forces are known as risk factors for issues and diseases especially of the lumbosacral joint L5-S1 (Bergmann et al., 2017). Various biomechanical models have been developed in order to predict the load on this joint. (e.g. Neumann, Wells and Norman, 1999), including the robust “Dortmunder” (Jäger and Luttmann, 1989). These models often work by means of manual data entry or automation with the help of so-called motion tracking systems (MTS). Nowadays, the measured movements and resulting forces at the joints can be visualized in real time. In the course of the project “Virtual Reality in Motion Lab” (VRiMoLa) we used a motion tracking system from the end user segment to feed data to a simplified biomechanical model. We also implemented force plates to measure additional forces resulting from added loads and inertia, as well as a VR-HMD (Head Mounted Display) to simultaneously visualize the data for the test subjects. The aim was to create a system to reliably output real-time joint load data in a digestible manner to be used in the study programs Health Sciences and Hazard Control at HAW Hamburg, teaching students about the dangers and remedial procedures in rescue and nursing activities. Methods: To validate the (simplified) model, students performed a nursing activity while wearing the tracking and VR hardware and standing on the force measuring plates. The results of the model calculations based on these data in regards to the joint L5-S1 were compared to results obtained in a study using the robust “Dortmunder” model. (Jäger et al., 2014). Results: 9 participants were equipped with the MTS and performed the nursing activity three times each. Each respective middle iteration was chosen for further analysis. Two of the 9 measurements analysed could not be used due to technical problems. In the study by Jäger et al. 2014, the same nursing activity was also performed in 7 iterations of the experiment. In the numerical comparison between our study and Jäger, the maxima of the pressure load on L5-S1 are very similar, while the mean was the same at 2.7 kN in both studies. Discussion: Even though the model used in the study was fairly simple, the results nonetheless correspond to the more rigorous Dortmunder model. Further development and validation is necessary to arrive at reliable model and surrounding software that can be employed in research- and teaching environments. Conclusion: To make use of the potential in real-time simulation, the necessary development of the calculations and an expansion of the scope, will be carried out in two dissertations, implementing a more robust model based on the Dortmunder, with added functionality for the forces not just in L5-S1, but also the shoulder.