Evaluation of stereo photogrammetry to assess full field vessel strain measurements following stent implantation

This work aimed to determine the accuracy of a stereo-photogrammetric technique for the full-field strain measurement employed in in-vitro analyses on stented vessels. To investigate the optical system an evaluation process based on a virtual reality environment was developed. The softwares and the support for this study were provided by the Department of Cardiovascular Science of the University of Sheffield. The evaluation procedure consisted of several steps involving the use of the finite element software ANSYS Mechanical, a modelling application capable of simulating physical objects (Realsoft 3D), different Matlab toolboxes and finally an image registration toolkit developed by the University of Sheffield called ShIRT. A virtual stereo-camera rig was reproduce using these softwares and the 2D images of the object of interest were captured by each camera. Then nodal coordinates of the surface were extracted employing the manual reconstruction or the image registration method and the strain distribution was derived. Finally the values of strain were compared to the data computed by ANSYS to assess the goodness of the measurement. To obtain a complete characterisation of the system a 2D strip, a 3D vessel and a 3D stented vessel geometries were defined and the material properties were varied to quantify the sensitivity to different tissue behaviour. Some preliminary tests were performed to examine the influence of the system parameters on the strain measure and fix the principle tests’ features. The results showed an error on strain of less than 0.01 in case of simple geometry. More complex geometries were affected by errors of the order of 0.02 which suggest a reliable accuracy for large deformation application. The tests also demonstrated that the image registration improved the strain agreement. The evaluation method has confirmed as an effective and convenient technique which can be predict the accuracy of photogrammetric strain measurements in advance. This application is a powerful tool for the in-vitro approach to reduce the effort in terms of time and funds, and could be used both in the academic and industrial research field.