An articulated spine and ribcage kinematic model for simulation of scoliosis deformities

Abstract

Musculoskeletal multibody modeling can offer valuable insight into aetiopathogenesis behind adolescent idiopathic scoliosis, which has remained unclear. However, the underlying model should represent anatomical joints with compatible kinematic constraints while allowing the model to attain scoliotic postures. This work presents an improved and kinematically determinate model including the whole spine and ribcage, which can attain typical scoliosis deformations of the thorax with compatible constraint strategy and simulate the interaction between all the bony segments of the ribcage and the spine. In the model, costovertebral/costotransverse joints were defined as universal joints based on reported anatomical studies.Articulations between ribs and the sternum were defined as spherical joints except in the ninth and tenth levels, which have one additional anteroposterior degree-of-freedom. The model is controlled by fifteen kinematic parameters including spinal rhythms and parameters relating to clinical metrics of scoliosis. These input values were measured from the bi-planar radiographs of a 17-year-old scoliosis patient with a right main thoracic curve of 33° Cobb angle. Dependent kinematic variables with clinical relevance were selected for validation purposes and compared with measurements from radiographs. The average errors of ribvertebra angles, rib-vertebra angle differences, and rib humps were 6.3° and 10.5°, and 8.7mm. The model appeared to reproduce the spine and rib deformation pattern conforming to radiographs, results in simulating the rib prominence, rib spread, rib-vertebra angles, and sternum orientation, therefore supporting the constraint definitions. The model can subsequently be used to investigate the kinetics of scoliosis and contribute to uncovering the aetiology.

Biomechanical Morphing for Personalized Fitting of Scoliotic Torso Skeleton Models

“…Note how the lack of soft tissue leads to interpenetrations at the ribs, which are eliminated when soft tissue is added. The method of (Shayestehpour et al, 2021) suffers similar interpenetrations, as evidenced in their results. By adding the soft tissue, we also increased significantly the fitting quality.…”

“…For one particular validation case, we used normal-dose radiographs of a patient from a previous study ( Shayestehpour et al, 2021 ), obtained with a digital Carestream DRX-Evolution system with automatic exposure control (see Patient 1 in Figure 2 ). For all other nine patients, we used the low-dose X-ray system described in Section 2.2 .…”

“…Low-dose X-rays offer beneficial properties for regular checkup of AIS patients, as they reduce the exposure to radiation. However, they come with challenges for the creation of computer models of the torso and hence the adoption of computational solutions for, (Shayestehpour et al, 2021), on the normal-dose case (patient 1). The columns indicate, from left to right, clinical metrics measured on radiographs, metrics obtained using (Shayestehpour et al, 2021), metrics with our method, and error improvement thanks to our method (positive values indicate that our method outperforms (Shayestehpour et al, 2021)).…”

“…Scoliosis is a 3D spinal deformity; therefore, correction in the sagittal plain is equally important as in the frontal one (Salmingo et al, 2014). For a more detailed description and definition of the metrics, the reader may refer to (Shayestehpour et al, 2021).…”

“…Then they improved their method, making landmark selection more user friendly and robust ( Vergari et al, 2018 ). Shayestehpour et al (2021) presented an articulated spine and rib cage kinematic model, which is able to attain scoliotic postures.…”