Introdução: A doença falciforme (DF) é a afecção hematológica hereditária de maior prevalência no mundo, sendo que a anemia falciforme (AF) é a forma mais grave com elevada morbidade e mortalidade. Alterações cardíacas são reconhecidamente associadas à AF, como aumento cavitário, hipertensão arterial pulmonar e disfunção diastólica do ventrículo esquerdo. No entanto, ainda não existe consenso quanto à função sistólica ventricular nesses pacientes. O objetivo deste estudo foi o de investigar a função ventricular de pacientes com anemia falciforme utilizando o strain e o twist ventricular obtidos pelo speckle tracking bidimensional, e reconhecer os indivíduos sob maior risco cardiovascular, em que a instituição precoce de tratamento específico poderá beneficiar essa população. Métodos: Foram recrutados 40 pacientes com anemia falciforme (23,5 ± 9,3 anos; 24 homens) e 40 controles saudáveis pareados por sexo e idade, submetidos à entrevista estruturada, ecocardiograma transtorácico, cintilografia pulmonar ventilação perfusão e coleta de amostras de sangue. Todos os indivíduos foram submetidos à avaliação ecocardiográfica convencional padrão e subsequente avaliação offline do strain sistólico biventricular e estudo rotacional do ventrículo esquerdo utilizando speckle-tracking ecocardiográfico. Resultados: Os pacientes com AF apresentaram volume do VE indexado, massa do VE indexado, pressão arterial pulmonar e E/E\' médio superiores aos controles. As medidas de FE do VE, excursão sistólica do plano da tricúspide (TAPSE), Strain Global do VE (longitudinal, circunferencial, radial) e Strain Global do VD (longitudinal) não evidenciaram diferenças entre os grupos. Twist VE se mostrou reduzido em comparação aos controles (7,4 ± 1,2? vs 10,7 ± 1,8? , P <0,0001 ) e Tempo de pico de rotação apical prolongado (366,7 ± 26,1ms vs 344,6 ± 11,7ms , P <0,0001 ). Algumas variáveis se mostraram fortemente relacionadas ao twist VE, como índice de gravidade clínico (Rho= - 0,97, Z value= - 6,05, P < 0,0001), relação E/E\' médio (r = 0,94, F valor= 156.9, p<0,0001), IVDFVE (índice do volume diastólico final do ventrículo esquerdo)(r = 0,81 e p<0,0001) e pressão sistólica arterial pulmonar (r = 0,72 e p<0,0001). Conclusões: Os resultados deste trabalho indicam que o twist ventricular esquerdo derivado do speckle tracking bidimensional encontra-se alterado em pacientes com anemia falciforme e função sistólica ventricular preservada, avaliada pela metodologia convencional, e existe forte correlação entre o twist ventricular esquerdo e o índice clínico de gravidade, relação E/E\', índice do volume diastólico final do ventrículo esquerdo e pressão sistólica arterial pulmonar. / Background: Sickle cell disease (SCD) is the most prevalent hematological condition in the world, with sickle cell anemia (SCA) being its most serious form, displaying a high level of morbidity and mortality. Cardiac changes are known to be associated with SCA, including an increase in cardiac chamber size, pulmonary hypertension and left ventricle diastolic dysfunction. However, there is still no consensus regarding the ventricular systolic function in these patients. The purpose of this study is to investigate the ventricular function of patients with sickle cell anemia utilizing the strain and ventricular twist, obtained by two-dimensional speckle tracking, as well as to identify individuals with higher cardiovascular risk, in which early application of specific treatment could benefit this group of people. Methods: 40 patients were recruited with sickle cell anemia (ages 23.5 ± 9.3 years; 24 males) and 40 healthy control individuals paired by gender and age, submitted to structured interviews, transthoracic echocardiogram, pulmonary scintigraphy and collection of blood samples. All individuals were submitted to a standard echocardiographic evaluation and subsequent off line evaluation of the biventricular systolic strain and rotational study of the left ventricle using echocardiographic speckle-tracking. Results: Patients with SCA presented LV volume indices, LV mass Indices, pulmonary arterial pressure and E/ E\' ratios statistically higher than the control individuals. Measurements of Ejection Fraction (EF) of the left ventricle, tricuspid annular plane systolic excursion (TAPSE), Overall LV Strain (Longitudinal, Circumferential, and Radial) and Overall RV Strain (Longitudinal) did not present differences between the groups. LV twist was significantly lower in relation to the control group (7,4 ± 1,2? vs 10,7 ± 1,8? , P <0,0001 ) and prolonged time to peak apical rotation (366.7 ± 26.1ms vs 344.6 ± 11.7ms , P <0.0001 ). In addition, some variables showed themselves to be strongly related to LV twist such as the clinical severity index ( Rho= - 0.97, Z value= - 6.05, P < 0.0001), E/E\' Ratio (r = 0.94, F value=156.9 e p<0.0001) ), left ventricle end diastolic volume index (LVEDV index) (r = 0,81, p<0,0001) and pulmonary systolic arterial pressure (r = 0.72 e p<0.0001). Conclusions: The results of this study indicate that the left ventricular twist derived from two-dimensional speckle tracking is altered in patients with sickle cell anemia and a preserved ventricular systolic function, evaluated using conventional methodology and that there is a strong correlation between left ventricular twist and the clinical severity index, E/E\' ratio, left ventricle end diastolic volume index and the pulmonary systolic arterial pressure.

[pt] Exames de ultrassom são uma ferramenta popular de aquisição de imagens na medicina atual por ser um procedimento não-invasivo, seguro e barato. Entretanto, inerente a qualquer exame de ultrassom encontra-se o ruído speckle, responsável pela degradação da imagem e dificultando tanto sua interpretação por parte de médicos e pacientes, quanto prejudicando a acurácia de métodos computacionais de pós processamento, como classificação, reconstrução, caracterização de tecidos e segmentação, entre outros. Portanto, métodos de remoção ou suavização deste ruído que preservem as principais características do conteúdo observado se fazem fundamentais para um avanço nestes processos. Definido como um ruído multiplicativo, que segue estatísticas não-Gaussianas e como fortemente correlacionado, sua solução ainda hoje é tema de debates e estudos. Neste trabalho apresentaremos diversos métodos de filtragem 2D que se propõem a reduzir ou solucionar o ruído speckle bem como métodos qualitativos para avaliar seus desempenhos e técnicas para escolher os melhores parâmetros de cada filtro a fim de eleger quais métodos melhor solucionam este ruído. / [en] Ultrasound exams are a popular tool for image acquisition in day-to-day medicine, since it is a noninvasive, safe and cheap procedure. However, speckle noise is intrinsic to any ultrasound exam, and it is responsible for image quality degradation and for hindering its interpretation by doctors and patients alike, while also impairing the accuracy of post processing computational methods, such as classification, reconstruction, tissue characterization and segmentation, among others. Hence, smoothing or denoising methods that preserves the observed content core attributes are essential for those processes. Defined as a multiplicative noise, following non-Gaussian statistics and as strongly correlated, its solution today is still a matter of debates and research. In this work, several 2D filters that aim to smooth or remove speckle noise along with qualitative methods to evaluate their performances and means of choosing their best parameters are presented.

L'objectif de cette thèse est de proposer et d'étudier une méthode de segmentation qui soit adaptée à l'imagerie échographique ultrasonore (US) et qui tienne compte de certaines contraintes rencontrées en milieu clinique. De ce fait, cette méthode se doit d'être robuste au bruit de speckle et à l'atténuation des ondes US dans le milieu, mais aussi rapide et ne nécessiter que peu, voire aucun paramètre à régler par l'opérateur. Dans ce cadre, les solutions fondées sur des contours actifs fondés sur la Minimisation de la Complexité Stochastique ont été étudiées. L'impact de différentes modélisations du speckle sur les résultats de ces techniques de segmentation a été caractérisé. Il a été montré qu'il est important de prendre en compte les variations de l'intensité moyenne du speckle induites par l'atténuation dans chaque région de l'image, à la fois pour la segmentation et pour l'analyse des propriétés du speckle. De plus, une stratégie hiérarchique de segmentation a été développée. Celle-ci permet notamment d'accroître la qualité des segmentations et de diminuer les temps de calcul.Les algorithmes de segmentation considérés étaient initialement conçus pour des formes polygonales peu adaptées à celles rencontrées dans le cadre d'applications médicales. Nous avons donc développé un nouveau modèle de contour fondé sur la théorie de l'information qui permet toujours une mise en oeuvre rapide des algorithmes et ne dépend d'aucun paramètre à régler par l'utilisateur. Testé sur des images synthétiques et réelles de fantômes échographiques, ce nouveau modèle permet de mieux décrire les formes régulières et arrondies des objets rencontrés en imagerie échographique. / The purpose of this PhD thesis is to propose and study a segmentation method adapted to echographic ultrasound imaging that could be clinically operational (i.e. fast and parameter-free) and robust to both the speckle noise and the attenuation of the ultrasonic signal in the medium. The solutions we studied rely on statistical active contour methods that are based on the Minimization of the Stochastic Complexity (MSC). The impact on the segmentation results of several speckle noise models that still lead to fast segmentation algorithms has been characterized. A key feature of these models, that appears to be crucial for both the segmentation and the speckle characterization, is the ability to take into account the spatial variation of the average intensity induced by the attenuation of the signal in the medium. In addition, we proposed a hierarchical optimization strategy that improves segmentation results and decreases the computation time.Finally, a novel contour model that is adapted to smooth boundaries that are met in medical imaging is also proposed for the considered MSC segmentation algorithms. The construction of this contour model relies on Information Theory concepts. It still allows one to get low computation times and does not contain any tuning parameter. Evaluations performed on synthetic images and real echographic phantom images indicate that this contour model provides better segmentation results for smooth inclusions that usually compose the echographic images.

This paper presents basic principles of optical coherence tomography, review of applications and basic categorization of these systems. Paper also deals with the typical properties of images from optical coherence tomography, especially speckle pattern. This paper also provides an overview of the origin of speckle noise and utilization of its dependence on microstructure of probed tissue for image classification based on textural analysis. Experimental part of this paper consists of phantom preparation, data acquisition by OCT system, implementation of speckle analysis in MATLAB and of testing of its functionality on standard textural dataset and also on acquired image phantom data. Speckle analysis is used for phantom image data segmentation.

Temperature drifts, tool deterioration, unknown vibrations as well as spindle play are major effects which decrease the achievable precision of computerized numerically controlled (CNC) lathes and lead to shape deviations between the processed work pieces. Since currently no measurement system exist for fast, precise and insitu 3d shape monitoring with keyhole access, much effort has to be made to simulate and compensate these effects. Therefore we introduce an optical interferometric sensor for absolute 3d shape measurements, which was integrated into a working lathe. According to the spindle rotational speed, a measurement rate of 2,500 Hz was achieved. In-situ absolute shape, surface profile and vibration measurements are presented. While thermal drifts of the sensor led to errors of several µm for the absolute shape, reference measurements with a coordinate machine show, that the surface profile could be measured with an uncertainty below one micron. Additionally, the spindle play of 0.8 µm was measured with the sensor.

Cell transplantation and stem cell therapy are promising approaches for regenerative medicine and are of interest to researchers and clinicians worldwide. However, currently, no imaging technique that allows three-dimensional in vivo inspection of therapeutically administered cells in host tissues is available. Therefore, we investigate magnetomotive optical coherence tomography (MM-OCT) of cells labeled with magnetic particles as a potential noninvasive cell tracking method. We develop magnetomotive imaging of mesenchymal stem cells for future cell therapy monitoring. Cells were labeled with fluorescent iron oxide nanoparticles, embedded in tissue-mimicking agar scaffolds, and imaged using a microscope setup with an integrated MM-OCT probe. Magnetic particle-induced motion in response to a pulsed magnetic field of 0.2 T was successfully detected by OCT speckle variance analysis, and cross-sectional and volumetric OCT scans with highlighted labeled cells were obtained. In parallel, fluorescence microscopy and laser speckle reflectometry were applied as two-dimensional reference modalities to image particle distribution and magnetically induced motion inside the sample, respectively. All three optical imaging modalities were in good agreement with each other. Thus, magnetomotive imaging using iron oxide nanoparticles as cellular contrast agents is a potential technique for enhanced visualization of selected cells in OCT.

Laser speckle contrast imaging (LSCI) and its extension, multi-exposure laser speckle contrast imaging (MELSCI) are non-invasive techniques to monitor peripheral blood perfusion. One of the main drawbacks of LSCI and MELSCI in clinical use is that the techniques are sensitive to tissue movement. Moreover, the image background contributes to unnecessary data. The aim of this project was to develop and evaluate different methods to detect local motion artifacts and image backgrounds in LSCI and MELSCI. In this project, three different methods were developed: one using statistical analysis and two using machine learning. The method based on classical statistics was developed in MATLAB with a dataset made up of 1797 frames of 256 x 320 images taken from a recording of a hand where the thumb and middle finger were taking turns making small movements while the middle finger was the subject of three different states made by an occlusion cuff (baseline, occlusion, and reperfusion). The main filter that was used in the first method was the Hampel filter. Furthermore, networks for the machine learning method were developed in Python using the same dataset but with 20,000 small patches extracted from the dataset of sizes 3 x 3 to 21 x 21 pixels. The first machine learning method was based on two-dimensional data patches, hence no time dimension was included, while the second machine learning method used three-dimensional data patches where the time dimension was included (from 1s to 10s). The generation of ground truth for the dataset was manually created frame by frame in a ground truth generation graphical user interface (GUI) in MATLAB. To assess the three methods, the Dice coefficient was used. The statistical method resulted in a Dice coefficient of 0.7557. The highest Dice coefficient for the machine learning method with a 2D dataset was 0.2902 (patch size 13 x 13) and the lowest was 0.2372 (patch size 7 x 7). For the machine learning method with 3D datasets, the patch size of 21 x 21 x 4 resulted in the highest Dice coefficient (0.5173), and the 21 x 21 x 40 model had the lowest Dice coefficient (0.1782). Since the two methods based on temporal data proved to be performing best in this project, one conclusion for further development of an improved model is the usage of temporal data in the training of a model. However, one important difference between the statistical method and the three-dimensional machine learning method is that the statistical method does not handle fast perfusion changes as well as the machine learning method and can not detect image background and static tissue. Therefore, the overall most useful method to further develop is the three-dimensional machine learning method.

No description available.

X-ray phase-contrast imaging has seen rapid development in recent decades due to its superior performance in imaging low-absorption objects, compared to traditional attenuation x-ray imaging. Having higher demand on coherence, x-ray phase-contrast imaging is performed mostly at synchrotrons. With the development of different imaging techniques, and the development of laboratory sources and x-ray optics, x-ray phase-contrast imaging can now be implemented on laboratory systems, which is promising and practical for broader range of applications. The subject of this thesis is the implementation, development and comparison of different laboratory phase-contrast methods using a liquid-metal-jet source. The three x-ray phase-contrast imaging methods included in this thesis are the propagation-, grating-, and speckle-based techniques. The grating-based method has been implemented on a laboratory system with a liquid-metal-jet source, which yields several times higher brightness than a standard solid-anode microfocus source. This allows shorter exposure time or a higher signal-to-noise ratio. The performance of the grating-based method has been experimentally and numerically compared with the propagation-based method, and the dose required to observe an object as a function of the object’s diameter has been investigated with simulations. The result indicates a lower dose requirement for the propagation-based method in this system but a potential advantage for the grating-based method to detect relatively large samples using a monochromatic beam. The speckle-based method, both the speckle-tracking and speckle-scanning techniques, has been implemented on a laboratory system for the first time, showing its adaptability to radiation of low temporal coherence. Tomography has been performed and shows the potential applications of this method on quantitative analysis on both absorption and phase information of materials. As a basis for further optimization and comparisons to other methods, the noise properties of the differential phase contrast of the speckle-based method have been studied and an analytical expression for the noise variance introduced, showing a similarity to the grating-based method. / Faskontrastavbildning med röntgenstrålning är en teknik som har utvecklats kraftigt de senaste årtiondena, eftersom den fungerar bättre än traditionella, absorptionsbaserade röntgenundersökningar för objekt med låg absorption. Den har dock höga krav på koherens, vilket gjort att den huvudsakligen används vid stora synkrotron-anläggningar. Tack vare utveckligen av nya avbildningstekniker, laboratoriekällor och röntgenoptik kan numera faskontrast användas även med laboratoriesystem, vilket är lovande då tekniken kan användas vid ett större antal olika tillämpningsområden Denna avhandling syftar till att tillämpa, utveckla och jämföra olika faskontrastmetoder i laboratoriemiljö, med en metallstråleröntgenkälla. De tre faskontrastmetoderna som behandlas i denna avhandling är propogation, gitter och speckelbaserad faskontrast. Den gitterbaserade metoden har implementerats i ett laboratoriesystem med en metallstrålekälla som ger flera gånger högre radians än en vanlig, fast mikrofokuskälla. Den högre radians en möjliggör kortare exponeringstider eller högre signal-brusförhållande. Den gitterbaserade tekniken har jämförts experimentellt och numeriskt med den propageringsbaserade metoden. Den strålningsdos som krävs för observera ett objekt, som funktion av dess diameter, har jämförts för de båda teknikerna, den här gången via simuleringar. Resultaten visar på en lägre strålningsdos för den propagationsbaserade tekniken i detta fall, men även att det finns en potentiell fördel för den gitterbaserade tekniken för något större objekt med monokromatisk röntgenstrålning. Speckelbaserade tekniker, nämare bestämt den som bygger på att spåra speckel och den som bygger på att scanna diffusorn, har för första gången implementerats i laboratoriemiljö. Därmed har visats att de fungerar även för strålning med låg tidskoherens. Tekniken har även använts för tomografi och visar möjliga tillämpningar inom kvantitativ analys av material. För att förenkla framtida optimeringar och jämförelser av tekniken med andra metoder, har brusegenskaperna för den speckelbaserade metoden studerats och visat sig likna den gitterbaserade metoden. / <p>QC 20160921</p>

Background: Coarctation of the aorta (CoA) is a congenital heart disease which represents a narrowing of the proximal descending aorta, hence increasing pressure afterload to the left ventricle (LV). Conventional treatment of native CoA is surgical repair, however potential recurrence or other related complications e.g. aortic rupture, heart failure and cerebrovascular events are common. Thus, lifelong follow-up of these patients is required. Echocardiography is the most patient’s friendly method to evaluate CoA and in particular its effect on LV function. Moreover, the novel speckle tracking echocardiography (STE) is an important method to assess subclinical LV dysfunction, a technique that promises better evaluation of LV function in these patients. The aims of this thesis were to review the literature on LV function in children with CoA using myocardial deformation imaging technologies, hence, to better understand the current knowledge and vagueness of the scientific evidence. We also aimed to study the effect of early CoA repair on the structure and function of LV and ascending aorta. In addition, we wished to establish in a meta-analysis format normal values of speckle tracking derived strain and strain rate values. Methods: Study 1. We have systematically searched the PubMed, and studies that fulfilled the inclusion criteria were critically analyzed and presented on a narrative form. Study 2 and 3. In addition to conventional echocardiographic measures of LV and ascending aorta, we measured longitudinal strain and strain rate of the LV using a vendor independent software, TomTec. We have also measured the aorto-septal angle (AoSA). Data was compared with normal healthy controls. Study 4. Electronic databases were systematically searched and suitable studies were meta analyzed using Comprehensive meta-analysis version 3 software. Results: Study 1. In 7/4945 included articles, 123 and 76 patients with congenital aortic stenosis (CAS) and CoA were reported, respectively. Normal conventional LV function, with subclinical myocardial dysfunction were reported in all studies before intervention. After intervention, a consistent improvement of myocardial deformation parameters was documented, even though not reaching normal values. Study 2. In 21 patients with CoA, LV function significantly improved after intervention (p &lt;0.001), however normal values were not reached even at medium-term follow-up (p = 0.002). Medium-term longitudinal strain correlated with pre intervention LV ejection faction (EF) (r = 0.58, p = 0.006). Medium-term subnormal values were more frequently associated with Bicuspid aortic valve (BAV) (33.3% vs. 66.6%; p &lt;0.05). Study 3. AoSA was abnormally wide before intervention, in particular at peak ejection in the descending aorta (p &lt;0.0001), and correlated with CoA pressure gradient. After intervention, AoSA normalized and significantly correlated with the increase of LV cavity function and overall LV deformation parameters. Study 4. In a meta-analysis of 28/282 studies including 1192 subjects, strain and strain rate values were established. Longitudinal strain normal mean values varied from -12.9 to -26.5 (mean, -20.5; 95 % CI, -20.0 to -21.0). Normal mean values of circumferential strain varied from -10.5 to -27.0 (mean, -22.06; 95 % CI, -21.5 to -22.5). Radial strain normal mean values varied from 24.9 to 62.1 (mean, 45.4; 95 % CI, 43.0 to 47.8). Meta-regression showed LV end-diastolic diameter as a significant determinant of variation of longitudinal strain. Longitudinal systolic strain rate was significantly determined by age and radial strain was influenced by the type of vendor used. Conclusion: The systematic review showed subclinical LV dysfunction in children with CoA before and after correction. However, since most of the patients were operated at an older age and had preserved LV EF, the effect of early intervention on LV function was only speculated. Our children with CoA who were operated at an earlier age showed LV subclinical dysfunction even at medium- term after intervention while the AoSA returned to normal shortly after intervention. Lower longitudinal strain values were found in patients with LV dysfunction (LV EF &lt;50%) before intervention and BAV. Finally, normal range values for strain and strain rate have been established and seem to be influenced by patients’ age, LV end-diastolic diameter and vendor used.