Anisotropic computational modelling of bony structures from CT data: An almost automatic procedure. (June 2020)

• Record Type: Journal Article
• Authors: Toniolo, Ilaria; Salmaso, Claudia; Bruno, Giovanni; De Stefani, Alberto; Stefanini, Cesare; Gracco, Antonio Luigi Tiberio; Carniel, Emanuele Luigi
• Is Part Of: Computer methods and programs in biomedicine. Volume 189(2020)
• Publication Date: 2020-06
• Subjects: Bone mechanics -- Orthotropic elasticity -- CT data -- Micromechanical modelling; Medicine -- Computer programs -- Periodicals; Biology -- Computer programs -- Periodicals; Computers -- Periodicals; Medicine -- Periodicals; Médecine -- Logiciels -- Périodiques; Biologie -- Logiciels -- Périodiques; Biology -- Computer programs; Medicine -- Computer programs; Periodicals; Electronic journals; 610.28
• DOI: 10.1016/j.cmpb.2020.105319
• Languages: English
• ISSNs: 0169-2607

Abstract

Highlights: Almost automatic procedure to the development of innovative computational clinical tools. Method of detection of the principal directions of anisotropy for the bone tissue. Identification of the principal factors and quantification of influence on the procedure's accuracy. Capability of customisation of the procedure. Abstract: Background and objective: The use of modelling techniques that combine CT data and bone tissue micromechanics is spreading in computational biomechanics. Finite Element models show great potential in surgical planning of intervention and in prediction of stress and strain fields through a non-invasive method. The main challenge pertains to the reliable characterization of bone mechanical behaviour. An almost automatic procedure is here defined, which provides computational models of bony structures considering the actual anisotropy of bone tissue response. The innovative aspect resides on the automatic detection of the directions of anisotropy as the eigenvectors of a three-dimensional distribution matrix of HU values. Methods: : The procedure combines CT data and micromechanics modelling techniques. Regarding a specific location, the procedure reports both the orthotropic elastic constants, by the analysis of the local HU value, and the anisotropic material directions, by the analysis of the HU values distribution around the specific location. Results: The procedure returns the distribution of bone tissue orthotropic elasticity tensor. The procedure proves to correctly respect the differentiation between cortical and trabecular bone. Principal directions show to be consistent with experimental data from ultrasound measurements. Regarding the material mapping from voxel to FE model, the developed strategies show to be reliable, leading to marginal errors (lower than 10%) for most of CT voxels (more than 90%). The computational analyses of typical structural loading conditions lead to strain values that are comparable with results from strain gauges experimentations. The development and the exploitation of FE models of different bony structures allow assessing the reliability of the procedure for cortical bone. Conclusions: The results highlight the potentialities of the procedure in providing accurate patient-specific biomechanical models of bony structures starting from CT data. The accuracy and the automatism of the procedure are important factors for the development of real time clinical tools. The main limitations of this work remain the not fully automatism and the reliability assessment, which is based mainly on cortical bone regions only.