Page | 43 Novel micro-CT based local strain mapping tool to characterize the failure modes of bone tissue engineering scaffolds G.Kerckhofs 1,2 , S. Van Bael 2,3 , M. Moesen 1,2 , F. Maes 4,5 , D. Loeckx 4,5 , J. Schrooten 1,2 , M. Wevers 1 1 Department of Metallurgy and Materials Engineering, K.U. Leuven, Kasteelpark Arenberg 44- bus 2450, B-3001 Heverlee, Belgium, greet.kerckhofs@mtm.kuleuven.be 2 Prometheus, Division of Skeletal Tissue Engineering, K.U.Leuven, B-3000 Leuven, Belgium 3 Division of Biomechanics and Engineering Design, K.U. Leuven , Celestijnenlaan 300C – bus 2419, B-3001 Heverlee, Belgium 4 Department of Electrical Engineering, Division ESAT - PSI, Centre for the Processing of Speech & Images, K.U. Leuven, Kasteelpark Arenberg 10 – bus 2441, B-3001 Heverlee, Belgium 5 Medical imaging centre, U.Z. Leuven, Herestraat 49 – bus 7003, B-3000 Leuven, Belgium Introduction. As proliferation and differentiation of osteogenic cells is also promoted by mechanical strains caused by the deformation of the structures to which the cells attach [1] , the local strain distribution inside bone tissue engineering (TE) scaffolds on which the cells are seeded can influence cell behaviour and subsequent bone formation. In order to evaluate bone TE scaffold designs in relation to this specific requirement, two approaches can be applied: (i) modelling by using finite element analysis (FEA) to predict the local strain distribution or (ii) experimental quantification of the local strain distribution. For the first, an accurate input of the material and the geometrical models is required as well as a thorough validation of the local strain predictions. For the latter, a combined use of micro-CT and in-situ loading can be applied, including experimental local strain mapping of the micro-CT scans at different loading steps. A novel local strain mapping approach was applied in this study based on non-rigid image registration [2] of the micro-CT scans at different loading steps. Figure 1. A typical (A) longitudinal 2D image of the CAD-model of design with beam length (L) 1.2 mm, (B) 3D CAD-model for the same beam length variation, (C) unit cell of the designed porous structures and (D) SLM fabricated open porous Ti6Al4V structure with designed beam length (L) 0.8 mm. Materials, methods and results. Open porous Ti6Al4V structures produced by rapid prototyping, and more specific selective laser melting (SLM), with varying beam length (L) and constant beam thickness were used as scaffold examples (fig. 1). The purpose of the local strain mapping was two-fold: (i) experimental quantification of the local strain distribution and failure modes in the selected scaffolds and (ii) comparison with the local strain distribution predicted by FEA. For the latter, to obtain input for the material model, bulk Ti6Al4V was produced by SLM and tested in compression. To optimize the geometrical model, the morphology of the produced structures was quantified via micro-CT imaging and 3D image analysis. C D B A L