Subject-specific 2D/3D image registration and kinematics-driven musculoskeletal model of the spine

An essential input to the musculoskeletal (MS) trunk models that estimate muscle and spine forces is kinematics of the thorax, pelvis, and lumbar vertebrae. While thorax and pelvis kinematics are usually measured via skin motion capture devices (with inherent errors on the proper identification of the underlying bony landmarks and the relative skin-sensor-bone movements), those of the intervening lumbar vertebrae are commonly approximated at fixed proportions based on the thorax-pelvis kinematics. This study proposes an image-based kinematics measurement approach to drive subject-specific (musculature, geometry, mass, and center of masses) MS models. Kinematics of the thorax, pelvis, and individual lumbar vertebrae as well as disc inclinations, gravity loading, and musculature were all measured via different imaging techniques. The model estimated muscle and lumbar forces in various upright and flexed postures in which kinematics were obtained using upright fluoroscopy via 2D/3D image registration. Predictions of this novel image-kinematics-driven model (Img-KD) were compared with those of the traditional kinematics-driven (T-KD) model in which individual lumbar vertebral rotations were assumed based on thorax-pelvis orientations. Results indicated that while differences between Img-KD and T-KD models remained small for the force in the global muscles (attached to the thoracic cage) (<15%), L4-S1 compression (<15%), and shear (<20%) forces in average for all the simulated tasks, they were relatively larger for the force in the local muscles (attached to the lumbar vertebrae). Assuming that the skin-based measurements of thorax and pelvis kinematics are accurate enough, the T-KD model predictions of spinal forces remain reliable.