Influence de l’intégration d’un modèle de la musculature cervicale active sur la cinématique tête-cou

Abstract

The cervical region is a particularly vulnerable anatomical area in road accidents, high-speed falls, and contact sports. The patterns of injury resulting from the mechanisms of injury depend on a wide variety of parameters. Among these factors, muscle contraction reduces the range of motion of the head and neck, thereby contributing to the stiffening of the cervical region. To elucidate these phenomena, numerical models of the head and neck incorporate Hill-type muscle contraction models. However, these same models have limited performance under high dynamic loads. The question that arises is therefore to what extent does the integration of cervical muscles using a refined Hill model improve the prediction of neck and head kinematics during a frontal impact? First, the open-source code for the refined Hill model (Extended Hill Type Model, EHTM) was translated and adapted for the OpenRadioss solver. An experimental muscle loading configuration on a porcine specimen was then reproduced in the finite element modelling software. The contraction speed results for four different loads (100g, 400g, 800g, 1600g) were compared with the experimental results, and a very good correlation was established for low loads with a CORA (Correlation Analysis) score of 0.89. In a second step, thirteen muscle groups were implemented in the SM2S (Spine model for safety and surgery) model and calibrated according to the physiological parameters of an anatomical study. The kinematic behaviour of the model was validated by reproducing a frontal impact test on a trolley with 2G deceleration conducted on ten volunteers. A sensitive study was set up to evaluate the influence of the activation delay and the level of muscle contraction activation on the kinematics of the head and neck. In addition, an innovative contraction method based on the physiological phenomenon of monosynaptic reflex was also evaluated. The study demonstrated that muscle activation parameters significantly influence head kinematics, with an opposite effect depending on the level of activation: rapid activation improves low-amplitude rotation (CORA = 0.82), while it deteriorates high-amplitude rotation (CORA = 0.36). However, the presence of several local maxima in the CORA score, sometimes outside physiologically plausible configurations, reveals a lack of robustness in the tabulated activation method. Despite still low CORA scores, the results obtained made it possible to account for the influence of the parameters of the tabulated activation pilot profiles for the cervical musculature on neck and head kinematics. This study constitutes the first validation data for a new Hill model at the muscle level on the OpenRadioss solver. Subsequent studies are planned to investigate the use of a closed-loop controller to pilot neural pilot and mimic conscious or unconscious muscle contraction without having to implement tabulated activation profiles.

Date	26 Aug 2025
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Marie-Hélène Beauséjour (Supervisor) & Éric Wagnac (Co-supervisor)