Development of a Multi-Modal Cervical Spine Motion Analysis System for Cervical Spondylosis Assessment

Abstract

Cervical spondylosis is typically assessed with global range-of-motion tests that miss level-specific dysfunction and do not quantify mechanical loading on the cervical spine. This project proposes a portable system with three inertial sensors at the forehead, C1, and C7 to capture segment-level kinematics during standardized movements and daily-task postures. Filtered orientation signals are resampled and segmented into clean trials, then converted to relative angles for upper (Head-C1) and lower (C1-C7) cervical segments, following established IMU protocols for cervical motion assessment and coupling analysis. A subject-scaled multibody inverse-dynamics model computes joint torques and axial loads from angular accelerations and gravity terms using anthropometric parameters; this bridges IMU kinematics with musculoskeletal kinetics that have been validated for spine applications. Moments and axial loads are mapped to bending and compressive stress/strain indices using level-specific geometry and finite-element-informed parameters, yielding normalized, clinician-readable risk bands. The device reports segmental ROM, asymmetry, coupling patterns, and load indices, and aggregates them into a severity score to classify normal, mild, moderate, or severe impairment. Prior work supports the validity of IMUs for cervical ROM and the feasibility of combining IMUs with musculoskeletal modeling to estimate spine kinetics, providing a foundation for the proposed diagnostic workflow. This level-aware, load-aware approach aims to improve bedside screening, targeted rehabilitation planning, and longitudinal monitoring in cervical spondylosis.