3D X-Ray imaging of bone tissue from micro to nano scale and associated inverse problems

F. Peyrin (1, 2) , Alina Toma (1) , B. Sixou (1) , L. Denis (3) , L. Wang (4) , L. Weber (2, 1) , M. Langer (2, 1) , P. Cloetens (2)

1 Imagerie Tomographique et Radiothérapie
2 ESRF - European Synchrotron Radiation Facility
3 LHC - Laboratoire Hubert Curien
4 MOTIVATE - Imagerie et modélisation Vasculaires, Thoraciques et Cérébrales

Imaging bone tissue from the organ to the cellular level is a major goal in bone research to understand, diagnose and predict bone fragility associated to bone disease such as osteoporosis. In this presentation, we show that X-ray CT is particularly well adapted to image bone in 3D up to the nanometer scale. After recalling the principles of 3D CT, we describe advances in bone CT imaging and the needs in associated inverse problems. Clinical X-ray CT is daily used to image skeletal tissue at the organ scale with a spatial resolution of about 0.5mm. However such systems do not permit to image bone micro-architecture made of a complex network of thin trabeculae (thickness about 150 µm). Imaging trabecular bone has been a driving application in the development of X-ray micro-Computerized Tomography (CT) ex-vivo. New High Resolution peripheral Quantitative CT (HR pQCT) systems provide images at voxel size around 100 µm, permitting the investigation of bone micro-structure in vivo [1]. Synchrotron X-ray CT, in addition to an accurate analysis of bone microarchitecture, provides quantitative information about the degree of mineralization of bone [2]. Finally, exploiting X-ray phase contrast has permit to reach 3D imaging of bone samples up to 50nm by using the magnified phase nano CT setup developed at the ESRF [3]. Basic CT reconstruction relies on the inversion of the Radon Transform, which is conventionally done using the Filtered Back Projection algorithm. Recently compressive sensing (CS) methods have raised increasing interest in CT imaging [4]. These algorithms are generally based on the minimization of a functional including a prior term promoting some form of sparsity. While many progresses have been made in the development of methods and algorithms, their practical uses in applications still requires developments. We describe current works and perspective in this field concerning bone imaging.

Domaines: Traitement du signal et de l'image [eess.SP]

F. Peyrin, Alina Toma, B. Sixou, L. Denis, L. Wang, et al.. 3D X-Ray imaging of bone tissue from micro to nano scale and associated inverse problems. XXIIIème Congrès Général de la Société Française de Physique, Aug 2015, Strasbourg, France. ⟨hal-01285028⟩