Persistence of external trunk asymmetry after scoliosis treatment is frequent and difficult to predict by clinicians. This is a significant problem considering that correction of the apparent deformity is a major factor of satisfaction for patients.
This research project consists in the development of a physically based model of the human trunk in order to predict the external appearance after scoliosis surgery.
For each patient, a 3D pre and postoperative reconstruction of the bone structures was obtained from standard radiographs while the external geometry of the trunk was acquired by non-invasive 3D digitizers.
Three different tetrahedral layers were generated from the external surface of the trunk to represent the skin, fat and muscles. From these layers a generalized particle system based on elastic potential energy with respect to various constraints was defined. Knowing the preoperative and postoperative configurations of the bone structures, our system propagates the correction through the different layers by using an optimization algorithm to reach the equilibrium state of the previously defined particle system to provide the simulated external geometry of the trunk.
Clinical indices of torso asymmetry were computed and compared by using crosssections of the simulated and acquired postoperative external geometry at various vertebral levels. A preliminary evaluation study on one patient showed a mean error ±1.38°(thoracic) and ±3.26°(lumbar) on the BSR index.
Once validated, the proposed tool will allow the clinicians to simulate the external outcome of a spine surgery and choose a surgical strategy that takes into account the expectations of their patients.