Search

AbstractMagnetic medical imaging (MRI) is commonly used to 3D rendering of musculoskeletal structures. However, due to technical limitations of the device e.g. the coil, high-spacial resolution is available only for a small field of view. However in the case of hand and forearm investigation, and due to the long tendon and small bones, both high special and a large field of view are required. It is the aim of this study to propose the in silicoenlargement of the field of view to 3D visualise all hand’s bones and tendons paths. Using MRI clinical routine sequence, we detailed a procedure of hand immobilisation by alginate and registration of up to four consecutive MRI volume acquisitions. In order to validate our method, we evaluated the accuracy of the anatomical structures segmentation (i), the robustness of the registration by a sensitivity analysis (ii) and the accuracy of the 3D reconstruction by comparison with computer tomography imaging (iii). The results demonstrated a mean error of 1 mm on the 3D external surface and of 0.94 mm on the localisation of the hand bones. We also tested different hand sizes and showed that the field of view could be increased up to 2.3 time compared to the original one.

A robust, accurate and flexible computational protocol to create personalised finite element models of musculoskeletal systems from medical images is presented. A contour-based geometrical reconstruction well suited to the complex and intrinsic character of living biological tissues and structures is used. The 3D surface model is decomposed into a number of stripes considered as quasi-developable domains which can be, therefore, meshed in a 2D parametric space. Once all boundary domains have been meshed, nodes are projected on a Hermite bicubic patch model derived from the segmentation points. A specific procedure to handle branching structures is proposed. In a final step, a tetrahedron mesh is generated and the mechanical properties could be assigned. Case studies using magnetic resonance imaging and computed tomography data are carried out and demonstrated. Results of the protocol are presented and discussed.

Quantitative information on annulus fibrosus (AF) and nucleus pulposus (NP) is of interest in early diagnosis of degenerated intervertebral disc (IVD) disease. In this present study, we developed a practical framework to analyse and visualise three feature maps of the IVD: the T2 relaxation time, the apparent diffusion coefficient and the anisotropy. They are derived from advanced Magnetic Resonance Imaging sequences such as T2 mapping, diffusion-weighted and diffusion tensor sequences. These features were quantified using regression analysis with least squares and signal intensity-based direction-oriented methods. Based on these maps, a threshold-based geometrical computing procedure was developed to compute the geometrical repartition maps. As a first case study, in vivofeatures of AF and NP regions of the L4/5 IVD were analysed and visualised. Practical relevance of our developed framework for a clinical purpose is discussed. In fact, our framework allows useful information of IVD to be quantified and used for early diagnosis as well as for further modelling purpose.

This paper adresses the methodology used to model the knee joint in vivo from MRI images. The knee joint model obtained is subject specific. The paper presents all the treatment procedure: Geometrical acquisitions, Joint movement analysis, Meshing techniques and nonlinear finite element modeling with contact between the bones, the cartilage and the menisci. The model provides the contact pressure applied on the various components of the joint for one normal subject.