Abstract: BACKGROUND: Previous biomechanical studies on cervical spondylosis have mostly been limited to single data analysis, but dynamic analysis is now being used to prove the differences in movement patterns between healthy subjects and cervical spondylotic myelopathy patients with skeletal muscle disorders.
OBJECTIVE: To provide more comprehensive evidence to support the biomechanical study of the cervical paraspinal muscles in spinal cord cervical spondylosis by analyzing the cross-sectional area of the cervical paraspinal muscles in spinal cord cervical spondylosis versus healthy populations and by simulating the differences in cervical paraspinal muscle force and coordinated activation of the cervical paraspinal muscles.
METHODS: This prospective study included 30 participants, comprising 20 cases in the cervical spondylotic myelopathy group and 10 cases in the healthy control group. A Siemens MAGNETOM Verio 3.0 T MRI scanner was used to perform detailed scans on the participants. The scanning range extended from the first cervical vertebra to the second thoracic vertebra, covering all vertebrae, intervertebral discs, and paravertebral soft tissues. Spin-echo sequences and inversion recovery sequences were used for data processing. Imaging scan parameters were set with a slice thickness of 3 mm, a slice spacing of 1 mm, and an image matrix of 320×224, with the resulting image sequences exported. Subsequently, we used ImageJ software to select the anterior flexor muscles and posterior extensor muscles of the cervical spine for image analysis, measuring the cross-sectional area of the muscles. The polygon selection tool was used to outline the muscle boundaries, and the muscle area was calculated based on the original scale in the MRI images. Six inertial sensors were used to capture the head and neck movement trajectories, and the converted head and neck movement trajectory parameters were imported into the OpenSim virtual simulation system to establish a muscle simulation model. Based on the results of the detailed muscle scans, the muscle force parameters of the simulation muscle model were adjusted proportionally to fine-tune and train the muscle model. Finally, computer algorithms were used to statically optimize the cervical spine movement data captured by the inertial sensors, calculate muscle excitation, and analyze muscle coordination status.
RESULTS AND CONCLUSION: (1) There were no significant differences in sex and age between the cervical spondylotic myelopathy and healthy subjects 
(P > 0.05). (2) The results of MRI anchor measurement of cervical paraspinal muscles showed that the cross-sectional area of the sternocleidomastoid muscle in the healthy control group was significantly greater than that in the cervical spondylotic myelopathy group (t=-2.501, P=0.019). The cross-sectional area of the multifidus muscle in the healthy control group was significantly greater than that in the cervical spondylotic myelopathy group (t=-2.437, P=0.022). There were no significant differences in the other muscles between the two groups (P > 0.05). (3) Compared with the healthy population, patients with cervical spondylotic myelopathy exhibited more pronounced muscle atrophy in the anterior and lateral cervical muscles. (4) Parameters from OpenSim virtual muscle force simulation showed that muscle force was significantly reduced in the cervical spondylotic myelopathy group compared with the healthy population. (5) Muscle activation results showed that muscle synergistic activation was significantly lower in the cervical spondylotic myelopathy group than in the healthy control group, indicating that patients with cervical spondylotic myelopathy have impaired synergistic balance in the cervical muscle groups. (6) Differences in magnetic resonance imaging and head and neck movement were observed between patients with cervical spondylotic myelopathy and the healthy controls.

Key words: MRI, cervical spondylosis, cervical paraspinal muscles, cervical spondylotic myelopathy, biomechanics, muscle activation, OpenSim