Search

International audience; The state-of-the art for solving the nonlinear material decomposition problem in spectral computed tomogra-phy is based on variational methods, but these are computationally slow and critically depend on the particular choice of the regularization functional. Convolutional neural networks have been proposed for addressing these issues. However, learning algorithms require large amounts of experimental data sets. We propose a deep learning strategy for solving the material decomposition problem based on a U-Net architecture and a Sim2Real transfer learning approach where the knowledge that we learn from synthetic data is transferred to a real-world scenario. In order for this approach to work, synthetic data must be realistic and representative of the experimental data. For this purpose, numerical phantoms are generated from human CT volumes of the KiTS19 Challenge dataset, segmented into specic materials (soft tissue and bone). These volumes are projected into sinogram space in order to simulate photon counting data, taking into account the energy response of the scanner. The network is trained to decompose the materials in the projection domain after which we apply any conventional tomographic method to reconstruct the dierent material volumes. The proposed decomposition method is compared to a regularized Gauss-Newton (RGN) method on synthetic data, experimental phantom data and human thorax data.

Cone-beam computed tomography (CBCT) is increasely used and in particular in dental imaging. However, dental CT images suffer from artifacts due to truncated projections, cone beam geometry or the presence of metal inserts. Low-dose dental CBCT reconstruction is therefore a difficult problem and not well addressed in the literature. In this paper, we evaluate an iterative method allowing to introduce a priori volume information and a deep learning method on real dental CBCT acquisitions. The results show that the deep learning method is more efficient for low-dose tomography and can accurately reconstruct a volume with only 20% of the normal dose., La tomographie 3D avec faisceau conique (CBCT) s’est beaucoup développée dernièrement et en particulier en imagerie dentaire.Toutefois, les images dentaires tomodensitométriques souffrent d’artefacts dus aux projections tronquées, à la géométrique du faisceau conique ouencore à la présence de métal. La reconstruction CBCT dentaire en faible dose est donc un problème difficile et peu traité dans la littérature. Dansce papier, nous évaluons une méthode itérative permettant d’introduire des informations a priori sur le volume et une méthode d’apprentissageprofond sur des acquisitions réelles CBCT dentaires. Les résultats montrent que la méthode par apprentissage profond permet de reconstruire avecdavantage de précision un volume avec seulement 20% de la dose normale

International audience; 3D Computerized Tomography (CT) is a gold standard technique to assess bone microstructure in the context of bone diseases such as osteoporosis. However, when acquired invivo, bone images may suffer from a low spatial resolution and the presence of noise due to the limited tolerable radiation exposure. One way to overcome this issue consists in applying Super-Resolution (SR) techniques that aim at recovering high resolution images. Significant progress has been recently made thanks to deep learning SR methods trained on natural image datasets. To measure the reconstruction quality, Peak Signal to Noise Ratio (PSNR) and Structural Similarity (SSIM) are commonly used in the SR literature. In this paper, we give evidence of the limitation of these two criteria. Through extensive experiments performed from a dataset of mice tibias specifically collected and imaged for this study, we show that state of the art deep learning-based SR methods miss important details about the bone microstructure which is not reflected by the PSNR and SSIM values. This study opens the door to future promising lines of research including new SR methods regularized with respect to morphometric and topological parameters of bone microstructures.