The introduction of dual energy CT, DECT, in the field of medical healthcare has made it possible to extract more information of the scanned objects. This in turn has the potential to improve the accuracy in radiation therapy dose planning. One problem that remains before successful material decomposition can be achieved however, is the presence of beam hardening and scatter artifacts that arise in a scan. Methods currently in clinical use for removal of beam hardening often bias the CT numbers. Hence, the possibility for an appropriate tissue decomposition is limited. Here a method for successful decomposition as well as removal of the beam hardening artifact is presented. The method uses effective linear attenuations for the five base materials, water, protein, adipose, cortical bone and marrow, to perform the decomposition on reconstructed simulated data. This is performed inside an iterative loop together with the polychromatic x-ray spectra to remove the beam hardening

Brachytherapy is radiation therapy performed by placing a radiation source near or inside a tumor. Difference between the current water-based brachytherapy dose formalism (TG-43) and new model based dose calculation algorithms (MBSCAs) can differ by more than a factor of 10 in the calculated doses. There is a need for voxel-by-voxel cross-section assignment, ideally, both the tissue composition and mass density of every voxel should be known for individual patients. A method for determining tissue composition via three material decomposition (3MD) from dual energy CT scans was developed at Linköping university. The method (named DIRA) is a model based iterative reconstruction algorithm that utilizes two photon energies for image reconstruction and 3MD for quantitative tissue classification of the reconstructed volumetric dataset. This thesis has investigated the accuracy of the 3MD method applied on prostate tissue in an anthropomorphic phantom when using two different approximations of soft tissues in DIRA. Also the distributions of CT-numbers for soft tissues in a contemporary dual energy CT scanner have been determined. An investigation whether these distributions can be used for tissue classification of soft tissues via thresholding has been conducted. It was found that the relative errors of mass energy absorption coefficient (MEAC) and linear attenuation coefficient (LAC) of the approximated mixture as functions of photon energy were less than 6 \% in the energy region from 1 keV to 1 MeV. This showed that DIRA performed well for the selected anthropomorphic phantom and that it was relatively insensitive to choice of base materials for the approximation of soft tissues. The distributions of CT-numbers of liver, muscle and kidney tissues overlapped. For example a voxel containing muscle could be misclassified as liver in 42 cases of 100. This suggests that pure thresholding is insufficient as a method for tissue classification of soft tissues and that more advanced methods should be used.

State-of-the-art reconstruction algorithms for medical helical cone-beam Computed Tomography (CT) are of type non-exact Filtered Backprojection (FBP). They are attractive because of their simplicity and low computational cost, but they produce sub-optimal images with respect to artifacts, resolution, and noise. This thesis deals with possibilities to improve the image quality by means of iterative techniques. The first algorithm, Regularized Iterative Weighted Filtered Backprojection (RIWFBP), is an iterative algorithm employing the non-exact Weighted FilteredBackprojection (WFBP) algorithm [Stierstorfer et al., Phys. Med. Biol. 49, 2209-2218, 2004] in the update step. We have measured and compared artifact reduction as well as resolution and noise properties for RIWFBP and WFBP. The results show that artifacts originating in the non-exactness of the WFBP algorithm are suppressed within five iterations without notable degradation in terms of resolution versus noise. Our experiments also indicate that the number of required iterations can be reduced by employing a technique known as ordered subsets. A small modification of RIWFBP leads to a new algorithm, the Weighted Least Squares Iterative Filtered Backprojection (WLS-IFBP). This algorithm has a slightly lower rate of convergence than RIWFBP, but in return it has the attractive property of converging to a solution of a certain least squares minimization problem. Hereby, theory and algorithms from optimization theory become applicable. Besides linear regularization, we have examined edge-preserving non-linear regularization.In this case, resolution becomes contrast dependent, a fact that can be utilized for improving high contrast resolution without degrading the signal-to-noise ratio in low contrast regions. Resolution measurements at different contrast levels and anthropomorphic phantom studies confirm this property. Furthermore, an even morepronounced suppression of artifacts is observed. Iterative reconstruction opens for more realistic modeling of the input data acquisition process than what is possible with FBP. We have examined the possibility to improve the forward projection model by (i) multiple ray models, and (ii) calculating strip integrals instead of line integrals. In both cases, for linearregularization, the experiments indicate a trade off: the resolution is improved atthe price of increased noise levels. With non-linear regularization on the other hand, the degraded signal-to-noise ratio in low contrast regions can be avoided. Huge input data sizes make experiments on real medical CT data very demanding. To alleviate this problem, we have implemented the most time consuming parts of the algorithms on a Graphics Processing Unit (GPU). These implementations are described in some detail, and some specific problems regarding parallelism and memory access are discussed.

High-resolution micro computed tomography (micro-CT) offers 3D image resolution of 1 um for non-destructive evaluation of various samples. However, the micro-CT performance is limited by several factors. Primarily, scan time is extremely long, and sample dimension is restricted by the x-ray beam and the detector size. The latter is the cause for the well-known interior problem. Recent advancement in image reconstruction, spurred by the advent of compressed sensing (CS) theory in 2006 and interior tomography theory since 2007, offers great reduction in the number of views and an increment in the volume of samples, while maintaining reconstruction accuracy. Yet, for a number of reasons, traditional filtered back-projection based reconstruction methods remain the de facto standard on all manufactured scanners. This work demonstrates that CS based global and interior reconstruction methods can enhance the imaging capability of micro-CT scanners. First, CS based few-view reconstruction methods have been developed for use with data from a real micro-CT scanner. By achieving high quality few-view reconstruction, the new approach is able to reduce micro-CT scan time to up to 1/8th of the time required by the conventional protocol. Next, two new reconstruction techniques have been developed that allow accurate interior reconstruction using just a limited number of global scout views as additional information. The techniques represent a significant progress relative to the previous methods that assume a fully sampled global scan. Of the two methods, the second method uses CS techniques and does not place any restrictions on scanning geometry. Finally, analytic and iterative reconstruction methods have been developed for enlargement of the field of view for the interior scan with a small detector. The idea is that truncated projections are acquired in an offset detector geometry, and the reconstruction procedure is performed through the use of a weighting function / weighted iteration updates, and projection completion. The CS based reconstruction yields the highest image quality in the numerical simulation. Yet, some limitations of the CS based techniques are observed in case of real data with various imperfect properties. In all the studies, physical micro-CT phantoms have been designed and utilized for performance analysis. Also, important guidelines are suggested for future improvements. / Ph. D.

<p></p><p>One-sided ultrasonic non-destructive evaluation (UNDE) is extensively used to characterize structures that need to be inspected and maintained from defects and flaws that could affect the performance of power plants, such as nuclear power plants. Most UNDE systems send acoustic pulses into the structure of interest, measure the received waveform and use an algorithm to reconstruct the quantity of interest. The most widely used algorithm in UNDE systems is the synthetic aperture focusing technique (SAFT) because it produces acceptable results in real time. A few regularized inversion techniques with linear models have been proposed which can improve on SAFT, but they tend to make simplifying assumptions that show artifacts and do not address how to obtain reconstructions from large real data sets. In this thesis, we present two studies. The first study covers the model-based iterative reconstruction (MBIR) technique which is used to resolve some of the issues in SAFT and the current linear regularized inversion techniques, and the second study covers the direct deep learning (DDL) technique which is used to further resolve issues related to non-linear interactions between the ultrasound signal and the specimen.</p> <p>In the first study, we propose a model-based iterative reconstruction (MBIR) algorithm designed for scanning UNDE systems. MBIR reconstructs the image by optimizing a cost function that contains two terms: the forward model that models the measurements and the prior model that models the object. To further reduce some of the artifacts in the results, we enhance the forward model of MBIR to account for the direct arrival artifacts and the isotropic artifacts. The direct arrival signals are the signals received directly from the transmitter without being reflected. These signals contain no useful information about the specimen and produce high amplitude artifacts in regions close to the transducers. We resolve this issue by modeling these direct arrival signals in the forward model to reduce their artifacts while maintaining information from reflections of other objects. Next, the isotropic artifacts appear when the transmitted signal is assumed to propagate in all directions equally. Therefore, we modify our forward model to resolve this issue by modeling the anisotropic propagation. Next, because of the significant attenuation of the transmitted signal as it propagates through deeper regions, the reconstruction of deeper regions tends to be much dimmer than closer regions. Therefore, we combine the forward model with a spatially variant prior model to account for the attenuation by reducing the regularization as the pixel gets deeper. Next, for scanning large structures, multiple scans are required to cover the whole field of view. Typically, these scans are performed in raster order which makes adjacent scans share some useful correlations. Reconstructing each scan individually and performing a conventional stitching method is not an efficient way because this could produce stitching artifacts and ignore extra information from adjacent scans. We present an algorithm to jointly reconstruct measurements from large data sets that reduces the stitching artifacts and exploits useful information from adjacent scans. Next, using simulated and extensive experimental data, we show MBIR results and demonstrate how we can improve over SAFT as well as existing regularized inversion techniques. However, even with this improvement, MBIR still results in some artifacts caused by the inherent non-linearity of the interaction between the ultrasound signal and the specimen.</p> <p>In the second study, we propose DDL, a non-iterative model-based reconstruction method for inverting measurements that are based on non-linear forward models for ultrasound imaging. Our approach involves obtaining an approximate estimate of the reconstruction using a simple linear back-projection and training a deep neural network to refine this to the actual reconstruction. While the technique we are proposing can show significant enhancement compared to the current techniques with simulated data, one issue appears with the performance of this technique when applied to experimental data. The issue is a modeling mismatch between the simulated training data and the real data. We propose an effective solution that can reduce the effect of this modeling mismatch by adding noise to the simulation input of the training set before simulation. This solution trains the neural network on the general features of the system rather than specific features of the simulator and can act as a regularization to the neural network. Another issue appears similar to the issue in MBIR caused by the attenuation of deeper reflections. Therefore, we propose a spatially variant amplification technique applied to the back-projection to amplify deeper regions. Next, to reconstruct from a large field of view that requires multiple scans, we propose a joint deep neural network technique to jointly reconstruct an image from these multiple scans. Finally, we apply DDL to simulated and experimental ultrasound data to demonstrate significant improvements in image quality compared to the delay-and-sum approach and the linear model-based reconstruction approach.</p><br><p></p>

Avec l’augmentation du nombre de scanner et de la dose collective, le risque potentiel d’apparition d’effets stochastiques est accentué. Pour limiter au maximum ce risque, les principes de justification et d’optimisation doivent être appliqués avec rigueur. L’optimisation des pratiques a pour but de délivrer la dose la plus faible possible tout en conservant une qualité diagnostique des images. C’est une tâche complexe qui implique de trouver en permanence un compromis entre la dose délivrée et la qualité image résultante. Pour faciliter cette démarche, des évolutions technologiques ont été développées. Les deux évolutions majeures sont la modulation du courant du tube en fonction de l’atténuation du patient et l’apparition des reconstructions itératives (IR). L’introduction des IR a modifié les habitudes puisqu’elles permettent de conserver des indices de qualité image équivalents en réduisant les doses. Cependant, leurs utilisations s’accompagnent d’une modification de la composition et de la texture de l’image nécessitant d’utiliser des métriques adaptées pour les évaluer. Le but de cette thèse est d’évaluer l’impact d’une utilisation des IR sur la réduction de la dose et sur la qualité des images afin de proposer en routine pour tous les patients, des protocoles avec la dose la plus faible possible et une qualité image adaptée au diagnostic. La première partie de cette thèse est consacrée à une mise au point sur la problématique du compromis dose/qualité image en scanographie. Les métriques de qualité image et les indicateurs dosimétriques à utiliser, ainsi que le principe et l’apport des reconstructions itératives y sont exposés. La deuxième partie est consacrée à la description des trois étapes réalisées dans cette thèse pour atteindre les objectifs. La troisième partie est constituée d’une production scientifique de 7 articles. Le 1er article présente la méthodologie d’optimisation globale permettant la mise en place de protocoles Basses Doses en routine avec utilisation de niveaux modérés des IR. Le 2ème article évalue l’impact et l’apport sur la qualité des images obtenues pour des niveaux de doses très bas. Le 3ème et le 4ème article montrent l’intérêt d’adapter ou de proposer des protocoles optimisés selon la morphologie du patient. Enfin les 3 derniers articles, illustrent la mise en place de protocoles Très Basses Doses pour des structures ayant un fort contraste spontané. Pour ces protocoles les doses sont proches des examens radiographiques avec des niveaux élevés des IR. La démarche d’optimisation mise en place a permis de réduire considérablement les doses. Malgré une modification de la texture et de la composition des images, la qualité des images obtenues pour tous les protocoles était jugée satisfaisante pour le diagnostic par les radiologues. L’utilisation des IR en routine nécessite une évaluation particulière et un temps d’apprentissage pour les radiologues. / The increasing number of scanner and the cumulative dose delivered lead to potential risk of stochastic effects. To minimize this risk, optimization on CT usage should be rigorously employed. Optimization aims to deliver the lowest dose but maintaining image quality for an accurate diagnosis. This is a complex task, which requires setting up the compromise between the dose delivered and the resulting image quality. To achieve such goal, several CT technological evolutions have been developed. Two predominant developments are the Tube Current Modulation and the Iterative Reconstruction (IR). The former lays one patient's attenuation, the latter depend on advanced mathematical approaches. Using IR allows one to maintain equivalent image quality values by reducing the dose. However, it changes the composition and texture of the image and requires the use of appropriate metric to evaluate them. The aim of this thesis was to evaluate the impact of using IR on dose reduction and image quality in routine for all patients, protocols with the lowest dose delivered with an image quality suitable for diagnosis. The first part of the thesis addressed the compromise between dose delivered and image quality. Metrics of the image quality and the dosimetric indicators were applied as well the principle and the contribution of IRs were explored. The second part targets the description of the three steps performed in this thesis to achieve the objectives. The third part of the thesis consists of a scientific production of seven papers. The first paper presents the global optimization methodology for the establishment of low dose protocols in routine using moderate levels of IR. The second paper assesses the impact and contribution of IR to the image quality obtained to levels very low doses. The third and the fourth papers show the interest to adapt or propose protocols optimized according to patient's morphology. Finally the last three papers illustrate the development of Very Low Dose protocols for structures with high spontaneous contrast. For these protocols, doses are close to radiographic examinations with high levels of IR. The optimization process implementation has significantly doses reduction. Despite the change on the texture and on composition of the images, the quality of images obtained for all protocols was satisfactory for the diagnosis by radiologists. However, the use of routine IR requires special assessment and a learning time for radiologists.

Le scanner a révolutionné la médecine permettant une accélération et une meilleure prise en charge du patient. La tomodensitométrie (TDM) s’accompagne d’un désavantage qui est l’augmentation du risque de cancer radio-induit des patients qui en bénéficient. La question se pose notamment aux urgences où l’emploi du scanner est de plus en plus prédominant, souvent après la réalisation d’une radiographie. Cette attitude, malgré tout justifiée dans la plupart des cas, peut s'avérer délétère. De ce fait les principes de radioprotection obligent à l’optimisation de la dose délivrée aux patients. L’inquiétude principale réside dans l’irradiation du thorax qui est la région la plus radiosensible du corps humain. Cela se traduit par une recherche continue d’un compromis entre l’obtention de la dose la plus basse possible tout en gardant une qualité d’image satisfaisante pour le diagnostic. Les dernières années des innovations technologiques ont été développées pour optimiser la dose au scanner ; la plus importante et la plus récente étant la reconstruction itérative (RI). La RI permet d’améliorer les index de qualité image avec une dose abaissée ou à dose équivalente reconstruite avec la classique rétroprojection filtrée, mais restituant enfin une qualité d’image modifiée. L’objectif de cette thèse était d’établir un protocole TDM du thorax délivrant une dose similaire à celle d’une radiographie du thorax de face et une de profil (ULD-CT_Ultra-low-dose-Computed Tomography) pour des indications de douleurs thoraciques en urgence sans injection de produit de contraste. La réaction des radiologues non habitués a été investiguée pour considérer la modification de l’image liée à la réduction de la dose et de l’emploi de la RI. Pour atteindre cet objectif les travaux de cette thèse se sont déroulés selon trois phases. La première phase représente une approche globale à la RI, testée sur fantômes pour optimiser les protocoles TDM de notre département. À partir des résultats obtenus, la deuxième phase a débuté. Des protocoles TDM thorax standard, à basse dose (LD-CT_Low-dose) et à très basse dose (ULD-CT) ont été testés sur des cadavres humains. La troisième phase a été caractérisée par l’application du protocole ULD-CT en pratique clinique aux urgences. Quatre articles scientifiques ont été rédigés pour représenter les trois phases de cette thèse. En conclusion, le protocole ULD-CT reconstruit avec des hauts niveaux de RI a délivré une dose inférieure à celle du niveau de référence diagnostique national pour une radiographie du thorax de face et une de profil. Ce type de protocole à très faible dose reconstruit avec RI est une alternative valable à la radiographie pour certaines indications sélectionnées pour l’exploration du thorax en urgence. En outre les radiologues malgré des remarques critiques sur la qualité d’image de l’ULD-CT ont toujours déclaré un niveau de confiance diagnostique élevé. / Computed Tomography (CT) improved patients' health care. However CT has a major drawback, which is the ionizing irradiation of the patient with an ensuing radiation-induced cancer risk. This issue is particularly observed in emergency settings, where the CT is increasingly becoming a dominant tool for the care decision-making, often after a radiographic study. Although this attitude is justified in the majority of the cases, it could be deleterious. Thus the principles of radiation safety obligate to the optimization of radiation dose delivered to the patients. The main problem is that the chest is the most radiation sensitive region of the human body. Hence the research of the better trade-off between the dose reduction and a diagnostic image quality is mandatory. Recently, several technological improvements have been developed to optimize the radiation dose at CT. The newest and most important innovation is the iterative reconstruction (IR). IR improves the quality image indexes of a CT image generated with a lowered dose or equivalent to that reconstructed with filtered back projection. Finally this reconstruction method renders a modified CT image. The goals of this PhD thesis were: i) to establish an unenhanced CT protocol, delivering a dose in the range of a radiographic study (ULD_ultra-low-dose-CT), for chest pain indications with no need of contrast media administration and ii) to investigate the reaction of unaccustomed radiologists to ULD-CT imaging. To accomplish these tasks the work of this thesis has been split in three phases. In the first phase a study approaching globally the IR was carried out testing several CT protocols on phantoms, in order to optimize the CT protocols of our institution. The outcomes of this study opened the second phase. A standard dose CT, a low-dose-CT and an ULD-CT protocols were acquired on the chest of human cadavers. The third phase was characterized by the application of ULD-CT in clinical practice in emergency settings. Four scientific articles were produced to communicate the results of this doctorate work. In conclusion, the ULD-CT protocol, reconstructed with high strengths of IR, conveyed a dose lower than the one of the national diagnostic reference level for a double projections chest X-ray. This ULD-CT protocol with IR is a valid alternative to the radiography for the study of the chest, for selected indications in emergency settings. Moreover, despite the radiologists were censorious about the ULD-CT image quality, they demonstrated always a high diagnostic confidence level.

<p>Contemporary algorithms employed for reconstruction of 3D volumes from helical cone beam projections are so called non-exact algorithms. This means that the reconstructed volumes contain artifacts irrespective of the detector resolution and number of projection angles employed in the process. In this thesis, three iterative schemes for suppression of these so called cone artifacts are investigated.</p><p>The first scheme, iterative weighted filtered backprojection (IWFBP), is based on iterative application of a non-exact algorithm. For this method, artifact reduction, as well as spatial resolution and noise properties are measured. During the first five iterations, cone artifacts are clearly reduced. As a side effect, spatial resolution and noise are increased. To avoid this side effect and improve the convergence properties, a regularization procedure is proposed and evaluated.</p><p>In order to reduce the cost of the IWBP scheme, a second scheme is created by combining IWFBP with the so called ordered subsets technique, which we call OSIWFBP. This method divides the projection data set into subsets, and operates sequentially on each of these in a certain order, hence the name “ordered subsets”. We investigate two different ordering schemes and number of subsets, as well as the possibility to accelerate cone artifact suppression. The main conclusion is that the ordered subsets technique indeed reduces the number of iterations needed, but that it suffers from the drawback of noise amplification.</p><p>The third scheme starts by dividing input data into high- and low-frequency data, followed by non-iterative reconstruction of the high-frequency part and IWFBP reconstruction of the low-frequency part. This could open for acceleration by reduction of data in the iterative part. The results show that a suppression of artifacts similar to that of the IWFBP method can be obtained, even if a significant part of high-frequency data is non-iteratively reconstructed.</p>

Contemporary algorithms employed for reconstruction of 3D volumes from helical cone beam projections are so called non-exact algorithms. This means that the reconstructed volumes contain artifacts irrespective of the detector resolution and number of projection angles employed in the process. In this thesis, three iterative schemes for suppression of these so called cone artifacts are investigated. The first scheme, iterative weighted filtered backprojection (IWFBP), is based on iterative application of a non-exact algorithm. For this method, artifact reduction, as well as spatial resolution and noise properties are measured. During the first five iterations, cone artifacts are clearly reduced. As a side effect, spatial resolution and noise are increased. To avoid this side effect and improve the convergence properties, a regularization procedure is proposed and evaluated. In order to reduce the cost of the IWBP scheme, a second scheme is created by combining IWFBP with the so called ordered subsets technique, which we call OSIWFBP. This method divides the projection data set into subsets, and operates sequentially on each of these in a certain order, hence the name “ordered subsets”. We investigate two different ordering schemes and number of subsets, as well as the possibility to accelerate cone artifact suppression. The main conclusion is that the ordered subsets technique indeed reduces the number of iterations needed, but that it suffers from the drawback of noise amplification. The third scheme starts by dividing input data into high- and low-frequency data, followed by non-iterative reconstruction of the high-frequency part and IWFBP reconstruction of the low-frequency part. This could open for acceleration by reduction of data in the iterative part. The results show that a suppression of artifacts similar to that of the IWFBP method can be obtained, even if a significant part of high-frequency data is non-iteratively reconstructed.

A reconstruction algorithm in computerized tomography is a procedure for reconstructing the attenuation coefficientscient, a real-valued function associated with the object of interest, from the measured projection data. Generally speaking, reconstruction algorithms in CT fall into two categories: direct, e.g., filtered back-projection (FBP), or iterative. In this thesis, we discuss a new fast matrix-free iterative reconstruction method based on a polyenergetic model. While most modern x-ray CT scanners rely on the well-known filtered back-projection algorithm, the corresponding reconstructions can be corrupted by beam hardening artifacts. These artifacts arise from the unrealistic physical assumption of monoenergetic x-ray beams. In this thesis, to compensate, we use an alternative model that accounts for differential absorption of polyenergetic x-ray photons and discretize it directly. We do not assume any prior knowledge about the physical properties of the scanned object. We study and implement different solvers and nonlinear unconstrained optimization methods, such as a Newton-like method and an extension of the Levenberg-Marquardt-Fletcher algorithm. We explain how we can use the structure of the Radon matrix and the properties of FBP to make our method matrix-free and fast. Finally, we discuss how we regularize our problem by applying different regularization methods, such as Tikhonov and regularization in the 1-norm. We present numerical reconstructions based on the associated nonlinear discrete formulation incorporating various iterative optimization methods.