Dissertations / Theses on the topic 'CT quantification'

With the vast array of genetically altered (knockout) mice becoming available there is a need for quantitative, repeatable, and efficient methodologies to characterize the phenotypic consequences of knocking out specific genes. Since knockout animals often have the ability to compensate for a single missing gene, it is important to examine the structural, material and morphological properties to obtain a thorough understanding of the changes occurring. For this project, femurs of knockout mice were first scanned using microcomputed tomography (micro-CT) to obtain high-resolution images of the trabecular bone in the distal femur, as well as cortical bone in the mid-diaphysis. After scanning, the femurs were tested to destruction in four-point bending at the mid-diaphysis about the medial lateral axis of the femur. These methodologies allowed quantification of (1) morphologic properties such as bone volume fraction, trabecular properties and 2nd moment of the area (2) structural properties such as stiffness, maximum load at failure, and post yield deformation and (3) material properties such as bone mineral density, elastic modulus and yield strength. As part of two independent studies, two different knockout mice, cyclooxygenase-2 (COX-2 -/-) and Apolipoprotein E (APOE -/-), were examined for structure-function relationships using these methodologies. COX-2 knockout mice were found to have decreased mineral content in their femurs, and increased post yield deformation. APOE knockout mice at 10 weeks of age had decreased bone mass and structural properties. However, by 40 weeks of age APOE deficient mice caught up to and exceeded the structural properties and bone mass of their wild type counterparts.

Peptide receptor radionuclide therapy (PRRT) shows promising results in the treatment of neuroendocrine tumors. These tumors over-express somatostatin receptors that allow us to label somatostatin analogues with ¹⁷⁷Lu to deliver dose to the tumor. However, currently every patient receives the same amount of radioactivity of approximately 7400 MBq per treatment cycle. With this “one dose fits all” approach, differences between patients are not taken into account resulting in some being under-treated while others over-treated. The aim of this thesis was to develop a simple protocol for ¹⁷⁷Lu activity quantification for patients undergoing PRRT with the purpose of performing personalized dose assessments. Physics phenomena that influence image quantification were investigated. As electrons emitted in the decay of ¹⁷⁷Lu result in creation of Bremsstrahlung, Monte-Carlo simulations were performed to investigate this effect on image quantification of ¹⁷⁷Lu. Phantom experiments with different attenuation and scatter conditions were performed to test quantification accuracy and evaluate performance of several segmentation methods. Images were reconstructed using the OSEM algorithm and two scatter correction methods were compared. An experiment to measure the camera deadtime was performed by adding activity into a bottle placed in a cylindrical phantom. Plots for observed count rate vs. true count rate were made, and the deadtime was calculated based on the paralyzable model. The protocol was applied to four patient data, and OLINDA and voxelized dosimetry calculations were used to create dose volume histograms for the kidneys. Lastly, a graphical user interface to allow for the quantitative reconstruction of the data obtained using any of the manufacturers was developed. Our results suggest that Bremsstrahlung contributions to the detected energy spectrum in imaging studies of ¹⁷⁷Lu have no degrading effects in image quantification. Our protocol recovers the activity in kidneys to within 10%. The deadtime correction based on paralyzable model was accurate for the count rates measured. The deadtime corrections should be performed based on scatter corrected photopeak window instead of the full spectrum. Lastly, the dose delivered to the kidneys in patient data was lower than the suggested dose per session in order to reach current toxicity limits.
Science, Faculty of
Physics and Astronomy, Department of
Graduate

State-of-the-art imaging equipment has increased clinician's ability to make non-invasive diagnoses of coronary heart disease (CHD); however, high volumes of imaging data make manual abnormality detection cumbersome in practice. In addition, the interpretation of CTA heavily relies upon the previous knowledge of the clinician. These limitations have driven an intense research in the context of automated solutions for fast, reliable and accurate diagnosis. Accordingly, in this thesis, we present an automated framework for detection, localization and quantification of the non-calcified coronary plaques in cardiac computed tomography angiography (CTA). The first contribution of the thesis is a coronary segmentation algorithm that is adaptive to the contrast agent and employs a hybrid energy incorporating local and global image statistics in a segmentation framework using partial differential equations (PDEs). Accordingly, we illustrated with the help of experimental evidence that a volume-specific intensity threshold leads to an improved segmentation in CTA. In the subsequent step, we employed a hybrid region-based energy for improved segmentation in CTA imagery. The hybrid energy couples an intensity-based local term with an efficient discontinuity-based global model of the image for optimal segmentation. The proposed method is less sensitive to the local optima problem and helps in reducing false positives, as well as it allows a certain degree of freedom for the initialization. Moreover, we employed an auto-correction feature for improved segmentation, as an auto-corrected mask captures the emerging peripheries of the coronary tree during the curve evolution. The effectiveness of the proposed model is demonstrated with the help of both qualitative and quantitative results, with a mean accuracy of 80% across the CTA dataset. The capability to address the variations in initial mask and localization radii simultaneously, makes our algorithm a feasible choice for coronary segmentation. The second contribution of the thesis is an automatic approach to analyse the segmented coronary tree for the presence of non-calcified plaques. The specific focus of this work is detection of non-calcified plaques in CTA, as intensity overlap between blood, fat and non-calcified plaques make the detection challenging. Non-calcified plaques are identified based on mean radial profiles that average the image intensities in concentric rings around the vessel centreline. Subsequently, an SVM classifier is applied to differentiate the abnormal coronary segments from normal ones. A total of 32 CTA volumes have been analysed and a detection accuracy of 88.4% with respect to the manual expert has been achieved. For plaque-affected segments, we further proposed a derivative-based method to localize the position and length of the plaque inside the segment. The plaque localization accuracy has been around 83.2%. Moreover, the proposed model has been tested on three different CTA datasets and has produced consistent results, demonstrating its reproducibility for generic CTA data. The final contribution of the thesis is a method to segment and quantify the non-calcified plaque. After evaluating the vessel wall thickness, posterior probability based voxel classification has been performed to quantify the lumen and plaque, respectively. Both qualitative and quantitative results demonstrate that the proposed model shows a good agreement with three independent experts. To optimize the processing time, we employed sparse field method in a level-set based active contour evolution.

Les maladies coronariennes constituent l'ensemble des troubles affectant les artères coronaires. Elles sont la première cause mondiale de mortalité. Par conséquent, la détection précoce de ces maladies en utilisant des techniques peu invasives fournit un meilleur résultat thérapeutique, et permet de réduire les coûts et les risques liés à une approche interventionniste. Des études récentes ont montré que la tomodensitométrie peut être utilisée comme une alternative non invasive et fiable pour localiser et quantifier ces lésions. Cependant, l'analyse de ces examens, basée sur l'inspection des sections du vaisseau, reste une tâche longue et fastidieuse. Une haute précision est nécessaire, et donc seulement les cliniciens hautement expérimentés sont en mesure d'analyser et d'interpréter de telles données pour établir un diagnostic. Les outils informatiques sont essentiels pour réduire les temps de traitement et assurer la qualité du diagnostic. L'objectif de cette thèse est de fournir des outils automatisés de traitement d'angiographie CT, pour la visualisation et l'analyse des artères coronaires d'une manière non invasive. Ces outils permettent aux pathologistes de diagnostiquer et évaluer efficacement les risques associés aux maladies cardio-vasculaires tout en améliorant la qualité de l'évaluation d'un niveau purement qualitatif à un niveau quantitatif. Le premier objectif de ce travail est de concevoir, analyser et valider un ensemble d'algorithmes automatisés utiles pour la détection et la quantification de sténoses des artères coronaires. Nous proposons un nombre de techniques couvrant les différentes étapes de la chaîne de traitement vers une analyse entièrement automatisée des artères coronaires. Premièrement, nous présentons un algorithme dédié à l'extraction du cœur. L'approche extrait le cœur comme un seul objet, qui peut être utilisé comme un masque d'entrée pour l'extraction automatisée des coronaires. Ce travail élimine l'étape longue et fastidieuse de la segmentation manuelle du cœur et offre rapidement une vue claire des coronaires. Cette approche utilise un modèle géométrique du cœur ajusté aux données de l'image. La validation de l'approche sur un ensemble de 133 examens montre l'efficacité et la précision de cette approche. Deuxièmement, nous nous sommes intéressés au problème de la segmentation des coronaires. Dans ce contexte, nous avons conçu une nouvelle approche pour l'extraction de ces vaisseaux, qui combine ouvertures par chemin robustes et filtrage sur l'arbre des composantes connexes. L'approche a montré des résultats prometteurs sur un ensemble de 11 examens CT. Pour une détection et quantification robuste de la sténose, une segmentation précise de la lumière du vaisseau est cruciale. Par conséquent, nous avons consacré une partie de notre travail à l'amélioration de l'étape de segmentation de la lumière, basée sur des statistiques propres au vaisseau. La validation avec l'outil d'évaluation en ligne du challenge de Rotterdam sur la segmentation des coronaires, a montré que cette approche présente les mêmes performances que les techniques de l'état de l'art. Enfin, le cœur de cette thèse est consacré à la problématique de la détection et la quantification des sténoses. Deux approches sont conçues et évaluées en utilisant l'outil d'évaluation en ligne de l'équipe de Rotterdam. La première approche se base sur l'utilisation de la segmentation de la lumière avec des caractéristiques géométriques et d'intensité pour extraire les sténoses coronaires. La seconde utilise une approche basée sur l'apprentissage. Durant cette thèse, un prototype pour l'analyse automatisée des artères coronaires et la détection et quantification des sténoses a été développé. L'évaluation qualitative et quantitative sur différents bases d'examens cardiaques montre qu'il atteint le niveau de performances requis pour une utilisation clinique
Coronary heart diseases are the group of disorders that affect the coronary artery vessels. They are the world's leading cause of mortality. Therefore, early detection of these diseases using less invasive techniques provides better therapeutic outcome, as well as reduces costs and risks, compared to an interventionist approach. Recent studies showed that X-ray computed tomography (CT) may be used as an alternative to accurately locate and grade heart lesions in a non invasive way. However, analysis of cardiac CT exam for coronaries lesions inspection remains a tedious and time consuming task, as it is based on the manual analysis of the vessel cross sections. High accuracy is required, and thus only highly experienced clinicians are able to analyze and interpret the data for diagnosis. Computerized tools are critical to reduce processing time and ensure quality of diagnostics. The goal of this thesis is to provide automated coronaries analysis tools to help in non-invasive CT angiography examination. Such tools allow pathologists to efficiently diagnose and evaluate risks associated with CVDs, and to raise the quality of the assessment from a purely qualitative level to a quantitative level. The first objective of our work is to design, analyze and validate a set of automated algorithms for coronary arteries analysis with the final purpose of automated stenoses detection and quantification. We propose different algorithms covering different processing steps towards a fully automated analysis of the coronary arteries. Our contribution covers the three major blocks of the whole processing chain and deals with different image processing fields. First, we present an algorithm dedicated to heart volume extraction. The approach extracts the heart as one single object that can be used as an input masque for automated coronary arteries segmentation. This work eliminates the tedious and time consuming step of manual removing obscuring structures around the heart (lungs, ribs, sternum, liver...) and quickly provides a clear and well defined view of the coronaries. This approach uses a geometric model of the heart that is fitted and adapted to the image data. Quantitative and qualitative analysis of results obtained on a 114 exam database shows the efficiency and the accuracy of this approach. Second, we were interested to the problem of coronary arteries enhancement and segmentation. In this context, we first designed a novel approach for coronaries enhancement that combines robust path openings and component tree filtering. The approach showed promising results on a set of 11 CT exam compared to a Hessian based approach. For a robust stenoses detection and quantification, a precise and accurate lumen segmentation is crucial. Therefore, we have dedicated a part of our work to the improvement of lumen segmentation step based on vessel statistics. Validation on the Rotterdam Coronary Challenge showed that this approach provides state of the art performances. Finally, the major core of this thesis is dedicated to the issue of stenosis detection and quantification. Two different approaches are designed and evaluated using the Rotterdam online evaluation framework. The first approach get uses of the lumen segmentation with some geometric and intensity features to extract the coronary stenosis. The second is using a learning based approach for stenosis detection and stenosis. The second approach outperforms some of the state of the art works with reference to some metrics. This thesis results in a prototype for automated coronary arteries analysis and stenosis detection and quantification that meets the level of required performances for a clinical use. The prototype was qualitatively and quantitatively validated on different sets of cardiac CT exams

Work was undertaken in this thesis to assess the use of somatostatin image quantification with SPET and SPET-CT to aid the characterisation of disease within the body. Two radionuclide somatostatin analogues were used for this assessment, the first was NeoSPECT and the second, OctreoScan. The primary aim of work in this thesis was to assess the role of NeoSPECT imaging in the characterisation of disease within the lungs, that is, to differentiate benign from malignant disease. Two forms of image quantification were used in the NeoSPECT assessment, a tumour to background ratio (T:B) and a value of tumour percentage uptake (% uptake). Values of T:B and % uptake were calculated from SPET images acquired 2 hours post injection. T:B results from the benign group (n = 8) demonstrated a median T:B of 2.21, whilst the malignant group (n = 28) demonstrated a median T:B of 2.01. The differences between the groups were tested statistically via a Mann-Whitney test, which showed there to be no statistical difference between the groups (p=0.90, 95.4% CI of (-0.5598, 0.5498)). To undertake the calculation of % uptake a non-patient acquisition (a standard acquisition) was also required, unfortunately not all of the patient cohort used for the T:B assessment had this additional acquisition. As a result of this numbers were low for the % uptake assessment in patients with a benign histology (n = 2), therefore statistical analysis could not be performed. However, review of the range of values for each histology within the malignant group proved useful as no differences were demonstrated between the ranges of values which could help to differentiate between the histologies. Quantification of dual time point imaging was also assessed to determine if there were any variations in values calculated that could also help differentiate the different histologies. For this assessment patients were images at 2 and 4 hours post injection. Results from the Wilcoxon Signed Rank test of the T:B assessment found there to be no statistically significant difference between values of T:B calculated at 2 and 4 hours that was characteristic of tumour type (p=1.0 and p=0.14). The difference in % uptake between 2 and 4 hours was also assessed via a Wilcoxon Signed Rank test, this test also concluded there to be no significant difference value of % uptake between the two acquisitions of the malignant group (p = 0.73). An attempt was also made to quantify ‘other’ uptake within the mediastinum, however, a lack of anatomical information made correlation with histology impossible and as a result no firm conclusions relating image quantification to histology could be drawn from this work. Work from this thesis concluded no quantitative difference between tissue histology could be demonstrated using NeoSPECT, either from single or dual time point imaging. As a result of the NeoSPECT work a number of factors which limited the accuracy and reproducibility of SPET image quantification were identified. Towards the end of the NeoSPECT work hybrid imaging (SPET-CT) became available within the department at Glasgow Royal Infirmary. It was believed that hybrid imaging could resolve some of the limitations and subsequently improve the accuracy of SPET image quantification. However, NeoSPECT was removed from the market for a short period of time and therefore a similar somatostatin analogue, OctreoScan, was used to investigate if the accuracy of somatostatin image quantification could be improved as a result of SPET-CT and its associated reconstruction algorithms including a CT based attenuation correction. Firstly, a qualitative assessment of image quality using the new hybrid reconstructions techniques was undertaken via an observer study. Images were reconstructed with the existing reconstruction techniques, as used for the NeoSPECT work, and with the new hybrid imaging techniques. Four experienced observers blinded to reconstruction technique were asked to score images in terms of their overall image quality. A Friedman test was performed on the scores for each observer, three of the four observers demonstrated a statistically significant difference in their scores between the existing and new hybrid technique (p = 0.00, p = 0.003, p= 0.00), with the new hybrid technique being assigned the highest scores in terms of image quality. Images were also assessed semi-quantitatively via profile analysis which also demonstrated a clear differentiation between the existing and new hybrid techniques with increased image quality being demonstrated in the hybrid data set. The quantitative accuracy of hybrid imaging was also assessed using phantom data. For 111In the difference of the value of absolute activity calculated and that measured varied by 35% but this improved to 21 % when scatter and CT-attenuation based corrections were applied. For 99mTc a much more notable difference between the existing techniques used in chapter 2 and those available from the use of hybrid imaging was demonstrated, the difference in the value of absolute activity calculated and that measured improved from 67% to 0.04%, respectively. Work in this thesis clearly demonstrated an improvement in image quality and accuracy in SPET quantification as a result of hybrid imaging techniques.

This thesis presents an automated framework for quantitative vascular shape analysis of the coronary arteries, which constitutes an important and fundamental component of an automated image-based diagnostic system. Firstly, an automated vessel segmentation algorithm is developed to extract the coronary arteries based on the framework of active contours. Both global and local intensity statistics are utilised in the energy functional calculation, which allows for dealing with non-uniform brightness conditions, while evolving the contour towards to the desired boundaries without being trapped in local minima. To suppress kissing vessel artifacts, a slice-by-slice correction scheme, based on multiple regions competition, is proposed to identify and track the kissing vessels throughout the transaxial images of the CTA data. Based on the resulting segmentation, we then present a dedicated algorithm to estimate the geometric parameters of the extracted arteries, with focus on vessel bifurcations. In particular, the centreline and associated reference surface of the coronary arteries, in the vicinity of arterial bifurcations, are determined by registering an elliptical cross sectional tube to the desired constituent branch. The registration problem is solved by a hybrid optimisation method, combining local greedy search and dynamic programming, which ensures the global optimality of the solution and permits the incorporation of any hard constraints posed to the tube model within a natural and direct framework. In contrast with conventional volume domain methods, this technique works directly on the mesh domain, thus alleviating the need for image upsampling. The performance of the proposed framework, in terms of efficiency and accuracy, is demonstrated on both synthetic and clinical image data. Experimental results have shown that our techniques are capable of extracting the major branches of the coronary arteries and estimating the related geometric parameters (i.e., the centreline and the reference surface) with a high degree of agreement to those obtained through manual delineation. Particularly, all of the major branches of coronary arteries are successfully detected by the proposed technique, with a voxel-wise error at 0.73 voxels to the manually delineated ground truth data. Through the application of the slice-by-slice correction scheme, the false positive metric, for those coronary segments affected by kissing vessel artifacts, reduces from 294% to 22.5%. In terms of the capability of the presented framework in defining the location of centrelines across vessel bifurcations, the mean square errors (MSE) of the resulting centreline, with respect to the ground truth data, is reduced by an average of 62.3%, when compared with initial estimation obtained using a topological thinning based algorithm.

Aim of the PhD work was the development and validation of methods for the extraction of quantitative biomarkers from in vivo molecular imaging (from PET/CT oncological studies) to be correlated with ex vivo molecular imaging indexes (from hystological and proteomic data).

L'étude a été initiée à l'occasion de la disponibilité d'un algorithme itératif de reconstruction en tomographie par émission de positons, qui présente l'avantage d'atteindre la convergence grâce à une méthode de régularisation. Il a donc fallu évaluer ses performances et son éventuel apport en comparaison aux algorithmes de référence, puis étudier l'influence du seul paramètre accessible aux utilisateurs pour une optimisation en clinique. Pour cela, plusieurs tests ont été réalisés, sur fantômes d'abord, puis sur patients car les résultats obtenus n'étaient pas directement transposables en clinique. Cet algorithme a de nombreux avantages par rapport au standard actuel OSEM-MLEM (moins de bruit, meilleur contraste, meilleure détectabilité des lésions) mais pourrait encore être amélioré pour diminuer les artéfacts et la surestimation de certaines métriques, grâce à l'utilisation de fantômes plus anthropomorphiques et l'accès à plus de paramètres de reconstruction. Des travaux sont encore en cours avec l'éditeur
This study was initiated to evaluate an iterative reconstruction algorithm in positron emission tomography based on a regularization method to obtain convergence. Our aim was to assess its performance, in comparison with other currently available algorithms and to study the impact of the only parameter available to users for eventual optimization, both using anthropomorphic phantoms and clinical data. We confirm that this algorithm shows several advantages compared to the traditional OSEM-MLEM concerning noise, contrast and detectability. By using anthropomorphic phantoms and with access to more reconstruction parameters, the performance could be further improved to decrease the artefacts and the overestimation of certain metrics. Work in progress

Le syndrome de détresse respiratoire aiguë (SDRA) présente un taux de mortalité élevé, près de 40%, dans des unités de soins intensifs. Il est défini comme un ensemble de manifestations cliniques, radiologiques et physiologiques qui traduisent une intense inflammation pulmonaire et une hyperperméabilité pulmonaire, correspondant aux différentes agressions aiguës du poumon. Le prise en charge des patients atteints du SDRA nécessite une ventilation assistée qui, en cas de mauvaise adaptation des paramètres de ventilation, notamment, pression et volume, peut aggraver l'état du patient. Le réglage de ces paramètres est basé sur l'analyse de l'aération pulmonaire en réponse à la ventilation assistée. Cette analyse peut être faite sur des images tomodensitométriques (CT en anglais) après y avoir segmenté le parenchyme pulmonaire. Néanmoins, cette segmentation est entravée par l'augmentation de la densité du parenchyme, qui réduit le contraste entre le poumon et les structures voisines. Cette thèse cherche à fournir des outils de traitement d'images qui permettent aux experts l'analyse de l'aération pulmonaire dans des images CT acquises dans le cadre d'un projet sur le SDRA utilisant un modèle animal
Acute Respiratory Distress Syndrome (ARDS) is a life threatening disease presenting a high mortality of about 40% in intensive care units. It is the consequence of different pulmonary aggressions generating hypoxemia and pulmonary edema, which are radiologically expressed as infiltrations observable as opaque regions in the lung. The treatment of ARDS requires mechanical ventilation, which may deteriorate the state of the patient if the ventilation parameters, namely volume and pressure, are not correctly adjusted. To adjust the parameter settings to each individual case, lung aeration - in response to ventilation - needs to be assessed. This assessment can be done using computed tomography (CT) images. However, it requires the segmentation of the lung-parenchymal tissue, which is a challenging task in ARDS images due the opacities that hinder the image contrast. In this thesis we aim to provide the required tools for the experts to analyze the aeration in the images acquired within an ARDS project using an animal model

Idiopathic pulmonary fibrosis (IPF) is a progressive disease of unknown aetiology and with a high mortality rate. Treatment options are currently limited and the mean survival following diagnosis is 3 yrs. Recently, dynamic and static PET/CT has been used to image the IPF lung with the aim of improving the understanding of the disease mechanisms, monitoring disease progression and evaluating treatment effectiveness. Quantitative imaging of PET tracer uptake in the lung tissue of patients with any diffuse pulmonary disease is highly challenging due to the effects of breathing and the large pulmonary blood component. In this work, three main sources of error on PET image quantitation in the lung have been assessed and addressed. Firstly, the effect of air and blood within the lung tissue on PET parameter accuracy has been characterised and methods developed to reduce the uncertainties. Secondly, analysis of the effect of density variations in the lung, due to the respiratory cycle, has been determined and suggestions given for minimising the errors induced. Thirdly, methods for calculating the blood tracer uptake and volume within the lung have been established and used to assess the effect on quantitation in the lungs of IPF and control patients. Finally, a feasibility study has been performed using an improved PET analysis methodology to determine if more robust parameter estimates can be derived in the lung tissue of patients with diffuse lung diseases. Specifically, the new methodology was applied to patients with IPF who had undergone dynamic PET/CT studies with 18F-FDG and 18F-FMISO.

Thesis (M.S.); Submitted to Iowa State Univ., Ames, IA (US); 12 Dec 2003.
Published through the Information Bridge: DOE Scientific and Technical Information. "IS-T 2608" Maria Ambert Sanchez. 12/12/2003. Report is also available in paper and microfiche from NTIS.

Le cancer du sein est une pathologie fréquente pour lequel les examens TEP/TDM au ¹⁸F-FDG et IRM mammaire sont fréquemment réalisés en routine. Il existe cependant une sous-utilisation des informations apportées par chacune de ces techniques d'imagerie. En pratique, l’interprétation de ces examens est principalement basée sur l’analyse visuelle et l'analyse « quantitative » se résume généralement au SUVmax seul en TEP/TDM et à l’étude du rehaussement du signal après injection de produit de contraste en IRM mammaire (DCE-MRI). L’arrivée de nouvelles machines hybrides TEP/ IRM, nous a amené à évaluer l'apport d’une quantification avancée des images issues de chacune de ces modalités séparément et en combinaison. Cela rejoint un domaine en expansion « la radiomique » qui consiste à extraire un grand nombre de caractéristiques quantitatives des images médicales pour décrypter l’hétérogénéité tumorale ou améliorer la prédiction du pronostic.L’objectif de notre travail était d’étudier l’apport des données radiomiques extraites de l’imagerie TEP au ¹⁸F-FDG et de l’IRM avec injection de produit de contraste réalisées avant traitement pour caractériser l’hétérogénéité tumorale dans le cancer du sein, en prenant en compte les différents sous-types moléculaires de cancer du sein, à savoir les tumeurs luminales (Lum A, Lum B HER2- et Lum B HER2+), triple-négatives et HER2+. Une importance particulière a été portée sur la valeur prédictive des informations radiomiques extraites de ces 2 techniques d’imagerie pour prédire le pronostic dans un groupe de patientes traitées par chimiothérapie néo-adjuvante. L’influence de variations physiologiques telles que l’âge sur le calcul des données radiomiques dans le tissu mammaire normal et cancéreux séparément a également été explorée, de même que la variabilité multicentrique des index radiomiques. L’extraction de ces données radiomiques a été effectuée grace au logiciel LiFex développé au sein du laboratoire IMIV sur une base de données-patientes recueillie en rétrospective.Nous avons rapporté pour la première fois l’influence de l’âge sur le calcul des indices « radiomiques » en TEP dans le tissu mammaire sain dans 2 institutions différentes mais aussi dans les tumeurs mammaires notamment celle triple-négatives. Des associations significatives entre le « phénotype tumoral radiomique » en imagerie TEP et IRM et des données pronostiques reconnues dans le cancer du sein ont été mises en évidence. En outre, nous avons démontré l’existence d’une grande variabilité pour le « profil radiomique » en TEP parmi les tumeurs présentant le même sous-type moléculaire. Cela suggére l’existence d’informations non-redondantes au sein du « phénotype tumoral métabolique » de chaque tumeur mammaire défini par les données radiomiques. L’exploration de cette variabilité s’est révélée intéressante pour améliorer la prédiction de la réponse histologique chez les patientes avec des tumeurs triple-négatives traitées par chimiothérapie néo-adjuvante. Par ailleurs, les mesures effectuées dans la région mammaire péri-tumorale chez les patientes traitées par chimiothérapie néo-adjuvante se sont montrées prédictives pour les patientes avec des tumeurs Lum B HER2-. En IRM nous avons montré l’importance de standardiser la méthode de mesure des caractéristiques radiomiques. Nous avons observé que les caractéristiques radiomiques issues des images DCE-MRI étaient moins associées aux caractéristiques moléculaires des tumeurs et avaient une valeur prédictive moindre. Nous avons également proposé une nouvelle méthode relativement standardisée pour le calcul des données radiomiques en IRM mammaire avec des résultats intéressants mais cette méthode doit encore être optimisée. Cependant, nos résultats suggèrent que les données extraites de la totalité du volume tumorale en IRM compléteraient efficacement les caractéristiques radiomiques TEP et le sous-type moléculaire pour prédire la réponse à la chimiothérapie néo-adjuvante
Breast cancer is a common disease for which ¹⁸F-FDG PET/CT and breast MRI are frequently performed in routine practice. However, the different information provided by each of these imaging techniques are currently under-exploited. Indeed, in routine the interpretation of these scans is mainly based on visual analysis whereas the « quantitative » analysis of PET/CT data is generally limited to the sole use of the SUVmax while in breast MRI, simple parameters to characterize tumor enhancement after injection of contrast medium are used. The advent of PET/MRI machines, calls for an evaluation of the contribution of a more advanced quantification of each of the modalities separately and in combination in the setting of breast cancer. This is along with the concept of « Radiomics » a field currently expanding and which consists in extracting many quantitative characteristics from medical images used in clinical practice to decipher tumor heterogeneity or improve prediction of prognosis. The aim of our work was to study the contribution of radiomic data extracted from ¹⁸F-FDG PET and MRI imaging with contrast injection to characterize tumor heterogeneity in breast cancer taking into account the different molecular subtypes of breast cancer, namely luminal (Lum A, Lum B HER2- and Lum B HER2 +), triple-negative and HER2 + tumors. In this context, we focused on the prediction of prognosis in patients treated with neo-adjuvant chemotherapy. The influence of physiological variations such as age on the calculation of radiomic data in normal breast and breast tumors separately was also explored, as well as the multi-center variability of radioman features. Radiomic features were extracted using the LiFex software developed within IMIV laboratory. The patient database used for the studies were all retrospective data. We reported for the first time the influence of age on the values of radiomic features in healthy breast tissue in patients recruited from 2 different institutions but also in breast tumors especially those with a triple-negative subtype. Similarly, significant associations between the radiomic tumor phenotype in PET and MRI imaging and well-established prognostic factors in breast cancer have been identified. In addition, we showed a large variability in the PET « radiomic profile » of breast tumors with similar breast cancer subtype suggesting complementary information within their metabolic phenotype defined by radiomic features. Moreover, taking into account this variability has been shown to be of particular interest in improving the prediction of pathological response in patients with triple-negative tumors treated with neoadjuvant chemotherapy. A peri-tumoral breast tissue region satellite to the breast tumor was also investigated and appeared to bear some prognostic information in patients with Lum B HER2- tumors treated with neoadjuvant chemotherapy. In MR, we demonstrated the need to harmonize the methods for radiomic feature calculation. Overall, we observed that radiomic features derived from MR were less informative about the molecular features of the tumors than radiomic features extracted from PET data and were of lower prognostic value. Yet, the combination of the enhanced tumor volume in MR with a PET radiomic feature and the tumor molecular subtype yielded enhanced the accuracy with which response to neoadjuvant therapy could be predicted compared to features from one modality only or molecular subtype only

Emphysema, a subset of COPD, occurs due to an abnormal inflammatory response to noxious gases or particles leading an influx of immunologic cells. Recent studies have demonstrated endothelial dysfunction in COPD subjects and are suggestive of a vascular phenotype present in COPD that is not fully characterized. We hypothesize that processes affecting the pulmonary vasculature lead to early changes important in the pathogenesis of COPD. This work focuses on the use of multidetector-row computed tomography (MDCT)-based measures of pulmonary blood flow (PBF), mean transit time (MTT) and pulmonary vascular volume (TPVV) to gain new insights into vasculature-related changes present in COPD. As a precursor to using perfusion MDCT imaging to phenotype lung disease, we demonstrated good regional correlation of PBF measurements obtained with MDCT imaging and fluorescent microspheres (FMS) at a FMS piece size resolution of 1.9 cm3 and regional volume level of 8-10 cm3. Additionally, we developed an ex vivo perfusion system, and applied quantitative image analysis techniques to study the lung preparation's stability over 120 minutes. We further validated CT-based PBF and MTT measurements by demonstrating physiologically appropriate responses to a range of flow rates with this new system. Finally, quantitative MDCT-based measurements were used to characterize a novel phenotype of emphysema and test hypotheses regarding vasculature-related changes in smokers and COPD subjects. We demonstrated increased heterogeneity in regional MTT and PBF measurements in smokers with preclinical emphysema compared with smokers with normal lung function and imaging studies and nonsmokers. This data is supportive of the notion that inflammatory-based vascular responses to hypoxia are occurring in smokers susceptible to COPD, but are successfully blocked in smokers without signs of emphysema. A new CT-based measure, TPVV, was studied and we demonstrate its association with total lung volume and body size metrics. TPVV measurements correlated with measures of COPD severity. A trend linking increased TPVV with increased endothelial dysfunction was observed, suggesting that pathological changes of COPD have an effect on the pulmonary vasculature. This work demonstrates the importance of functional information that can compliment structural, anatomical information to answer questions based on the lung physiology and pathological disease processes.

Nowadays, lung cancer assessment is a complex and tedious task mainly per- formed by radiological visual inspection of suspicious pulmonary nodules, using computed tomography (CT) scan images taken to patients over time. Several computational tools relying on conventional artificial intelligence and computer vision algorithms have been proposed for supporting lung cancer de- tection and classification. These solutions mostly rely on the analysis of indi- vidual lung CT images of patients and on the use of hand-crafted image de- scriptors. Unfortunately, this makes them unable to cope with the complexity and variability of the problem. Recently, the advent of deep learning has led to a major breakthrough in the medical image domain, outperforming conven- tional approaches. Despite recent promising achievements in nodule detection, segmentation, and lung cancer classification, radiologists are still reluctant to use these solutions in their day-to-day clinical practice. One of the main rea- sons is that current solutions do not provide support to automatic analysis of the temporal evolution of lung tumours. The difficulty to collect and annotate longitudinal lung CT cases to train models may partially explain the lack of deep learning studies that address this issue. In this dissertation, we investigate how to automatically provide lung can- cer assessment through deep learning algorithms and computer vision pipelines, especially taking into consideration the temporal evolution of the pulmonary nodules. To this end, our first goal consisted on obtaining accurate methods for lung cancer assessment (diagnostic ground truth) based on individual lung CT images. Since these types of labels are expensive and difficult to collect (e.g. usually after biopsy), we proposed to train different deep learning models, based on 3D convolutional neural networks (CNN), to predict nodule malig- nancy based on radiologist visual inspection annotations (which are reasonable to obtain). These classifiers were built based on ground truth consisting of the nodule malignancy, the position and the size of the nodules to classify. Next, we evaluated different ways of synthesizing the knowledge embedded by the nodule malignancy neural network, into an end-to-end pipeline aimed to detect pul- monary nodules and predict lung cancer at the patient level, given a lung CT image. The positive results confirmed the convenience of using CNNs for mod- elling nodule malignancy, according to radiologists, for the automatic prediction of lung cancer. Next, we focused on the analysis of lung CT image series. Thus, we first faced the problem of automatically re-identifying pulmonary nodules from dif- ferent lung CT scans of the same patient. To do this, we present a novel method based on a Siamese neural network (SNN) to rank similarity between nodules, overpassing the need for image registration. This change of paradigm avoided introducing potentially erroneous image deformations and provided computa- tionally faster results. Different configurations of the SNN were examined, in- cluding the application of transfer learning, using different loss functions, and the combination of several feature maps of different network levels. This method obtained state-of-the-art performances for nodule matching both in an isolated manner and embedded in an end-to-end nodule growth detection pipeline. Afterwards, we moved to the core problem of supporting radiologists in the longitudinal management of lung cancer. For this purpose, we created a novel end-to-end deep learning pipeline, composed of four stages that completely au- tomatize from the detection of nodules to the classification of cancer, through the detection of growth in the nodules. In addition, the pipeline integrated a novel approach for nodule growth detection, which relies on a recent hierarchi- cal probabilistic segmentation network adapted to report uncertainty estimates. Also, a second novel method was introduced for lung cancer nodule classification, integrating into a two stream 3D-CNN the estimated nodule malignancy prob- abilities derived from a pre-trained nodule malignancy network. The pipeline was evaluated in a longitudinal cohort and the reported outcomes (i.e. nodule detection, re-identification, growth quantification, and malignancy prediction) were comparable with state-of-the-art work, focused on solving one or a few of the functionalities of our pipeline. Thereafter, we also investigated how to help clinicians to prescribe more accurate tumour treatments and surgical planning. Thus, we created a novel method to forecast nodule growth given a single image of the nodule. Partic- ularly, the method relied on a hierarchical, probabilistic and generative deep neural network able to produce multiple consistent future segmentations of the nodule at a given time. To do this, the network learned to model the mul- timodal posterior distribution of future lung tumour segmentations by using variational inference and injecting the posterior latent features. Eventually, by applying Monte-Carlo sampling on the outputs of the trained network, we esti- mated the expected tumour growth mean and the uncertainty associated with the prediction. Although further evaluation in a larger cohort would be highly recommended, the proposed methods reported accurate results to adequately support the ra- diological workflow of pulmonary nodule follow-up. Beyond this specific appli- cation, the outlined innovations, such as the methods for integrating CNNs into computer vision pipelines, the re-identification of suspicious regions over time based on SNNs, without the need to warp the inherent image structure, or the proposed deep generative and probabilistic network to model tumour growth considering ambiguous images and label uncertainty, they could be easily appli- cable to other types of cancer (e.g. pancreas), clinical diseases (e.g. Covid-19) or medical applications (e.g. therapy follow-up).
Avui en dia, l’avaluació del càncer de pulmó ´es una tasca complexa i tediosa, principalment realitzada per inspecció visual radiològica de nòduls pulmonars sospitosos, mitjançant imatges de tomografia computada (TC) preses als pacients al llarg del temps. Actualment, existeixen diverses eines computacionals basades en intel·ligència artificial i algorismes de visió per computador per donar suport a la detecció i classificació del càncer de pulmó. Aquestes solucions es basen majoritàriament en l’anàlisi d’imatges individuals de TC pulmonar dels pacients i en l’ús de descriptors d’imatges fets a mà. Malauradament, això les fa incapaces d’afrontar completament la complexitat i la variabilitat del problema. Recentment, l’aparició de l’aprenentatge profund ha permès un gran avenc¸ en el camp de la imatge mèdica. Malgrat els prometedors assoliments en detecció de nòduls, segmentació i classificació del càncer de pulmó, els radiòlegs encara són reticents a utilitzar aquestes solucions en el seu dia a dia. Un dels principals motius ´es que les solucions actuals no proporcionen suport automàtic per analitzar l’evolució temporal dels tumors pulmonars. La dificultat de recopilar i anotar cohorts longitudinals de TC pulmonar poden explicar la manca de treballs d’aprenentatge profund que aborden aquest problema. En aquesta tesi investiguem com abordar el suport automàtic a l’avaluació del càncer de pulmó, construint algoritmes d’aprenentatge profund i pipelines de visió per ordinador que, especialment, tenen en compte l’evolució temporal dels nòduls pulmonars. Així doncs, el nostre primer objectiu va consistir a obtenir mètodes precisos per a l’avaluació del càncer de pulmó basats en imatges de CT pulmonar individuals. Atès que aquests tipus d’etiquetes són costoses i difícils d’obtenir (per exemple, després d’una biòpsia), vam dissenyar diferents xarxes neuronals profundes, basades en xarxes de convolució 3D (CNN), per predir la malignitat dels nòduls basada en la inspecció visual dels radiòlegs (més senzilles de recol.lectar). A continuació, vàrem avaluar diferents maneres de sintetitzar aquest coneixement representat en la xarxa neuronal de malignitat, en una pipeline destinada a proporcionar predicció del càncer de pulmó a nivell de pacient, donada una imatge de TC pulmonar. Els resultats positius van confirmar la conveniència d’utilitzar CNN per modelar la malignitat dels nòduls, segons els radiòlegs, per a la predicció automàtica del càncer de pulmó. Seguidament, vam dirigir la nostra investigació cap a l’anàlisi de sèries d’imatges de TC pulmonar. Per tant, ens vam enfrontar primer a la reidentificació automàtica de nòduls pulmonars de diferents tomografies pulmonars. Per fer-ho, vam proposar utilitzar xarxes neuronals siameses (SNN) per classificar la similitud entre nòduls, superant la necessitat de registre d’imatges. Aquest canvi de paradigma va evitar possibles pertorbacions de la imatge i va proporcionar resultats computacionalment més ràpids. Es van examinar diferents configuracions del SNN convencional, que van des de l’aplicació de l’aprenentatge de transferència, utilitzant diferents funcions de pèrdua, fins a la combinació de diversos mapes de característiques de diferents nivells de xarxa. Aquest mètode va obtenir resultats d’estat de la tècnica per reidentificar nòduls de manera aïllada, i de forma integrada en una pipeline per a la quantificació de creixement de nòduls. A més, vam abordar el problema de donar suport als radiòlegs en la gestió longitudinal del càncer de pulmó. Amb aquesta finalitat, vam proposar una nova pipeline d’aprenentatge profund, composta de quatre etapes que s’automatitzen completament i que van des de la detecció de nòduls fins a la classificació del càncer, passant per la detecció del creixement dels nòduls. A més, la pipeline va integrar un nou enfocament per a la detecció del creixement dels nòduls, que es basava en una recent xarxa de segmentació probabilística jeràrquica adaptada per informar estimacions d’incertesa. A més, es va introduir un segon mètode per a la classificació dels nòduls del càncer de pulmó, que integrava en una xarxa 3D-CNN de dos fluxos les probabilitats estimades de malignitat dels nòduls derivades de la xarxa pre-entrenada de malignitat dels nòduls. La pipeline es va avaluar en una cohort longitudinal i va informar rendiments comparables a l’estat de la tècnica utilitzats individualment o en pipelines però amb menys components que la proposada. Finalment, també vam investigar com ajudar els metges a prescriure de forma més acurada tractaments tumorals i planificacions quirúrgiques més precises. Amb aquesta finalitat, hem realitzat un nou mètode per predir el creixement dels nòduls donada una única imatge del nòdul. Particularment, el mètode es basa en una xarxa neuronal profunda jeràrquica, probabilística i generativa capaç de produir múltiples segmentacions de nòduls futurs consistents del nòdul en un moment determinat. Per fer-ho, la xarxa aprèn a modelar la distribució posterior multimodal de futures segmentacions de tumors pulmonars mitjançant la utilització d’inferència variacional i la injecció de les característiques latents posteriors. Finalment, aplicant el mostreig de Monte-Carlo a les sortides de la xarxa, podem estimar la mitjana de creixement del tumor i la incertesa associada a la predicció. Tot i que es recomanable una avaluació posterior en una cohort més gran, els mètodes proposats en aquest treball han informat resultats prou precisos per donar suport adequadament al flux de treball radiològic del seguiment dels nòduls pulmonars. Més enllà d’aquesta aplicació especifica, les innovacions presentades com, per exemple, els mètodes per integrar les xarxes CNN a pipelines de visió per ordinador, la reidentificació de regions sospitoses al llarg del temps basades en SNN, sense la necessitat de deformar l’estructura de la imatge inherent o la xarxa probabilística per modelar el creixement del tumor tenint en compte imatges ambigües i la incertesa en les prediccions, podrien ser fàcilment aplicables a altres tipus de càncer (per exemple, pàncrees), malalties clíniques (per exemple, Covid-19) o aplicacions mèdiques (per exemple, seguiment de la teràpia).

This project has investigated how the architecture of the blood vessels supplying nutrients to skeletal muscles is affected by muscle contusion injuries, and how it changes during healing with or without initial treatment of the injury by icing. In order to do this, we used contrast agents to visualise blood vessels in 3D with micro-computed tomography imaging. This research significantly contributes to the fields of orthopaedics, traumatology and sports medicine, as it improves our understanding of muscle contusion injuries. Furthermore, the methods developed in this thesis may help to improve the diagnosis and monitoring of these injuries.

Background To investigate vascular-related pathophysiological characteristics of two human lung cancers with modifiable vascularisation using MRI and CT. Methods Tumour xenografts with modifiable vascularisation were established in 71 rats (approval by the Animal Care Committee was obtained) by subcutaneous transplantation of two human non-small-cell lung cancer (NSCLC) cells (A549, H1299) either alone or co-transplanted with vascular growth promoters. The vascularity of the tumours was assessed noninvasively by MRI diffusion-weighted-imaging (DWI), T2-weighted, and time-of-flight (TOF) sequences) as well as contrast-enhanced CT (CE-CT), using clinical scanners. As a reference standard, histological examinations (CD-31, fluorescent beads) were done after explantation. Results Microvessel density (MVD) was higher in co-transplanted tumours (171 ± 19 number/mm2) than in non-co-transplanted tumours (111 ± 11 number/mm2; p = 0.002). Co-transplanted tumours showed higher growth rates and larger tumour vessels at TOF-MRI as well as larger necrotic areas at CE-CT. In co-transplanted tumours, DWI revealed higher cellularity (lower minimal ADCdiff 166 ± 15 versus 346 ± 27 mm2/s × 10−6; p < 0.001), highly necrotic areas (higher maximal ADCdiff 1695 ± 65 versus 1320 ± 59 mm2/s × 10−6; p < 0.001), and better-perfused tumour stroma (higher ADCperf 723 ± 36 versus 636 ± 51 mm2/s × 10−6; p = 0.005). Significant correlations were found using qualitative and quantitative parameters: maximal ADCperf and MVD (r = 0.326); maximal ADCdiff and relative necrotic volume on CE-CT (r = 0.551); minimal ADCdiff and MVD (r = −0.395). Conclusions Pathophysiological differences related to vascular supply in two human lung cancer cell lines with modifiable vascularity are quantifiable with clinical imaging techniques. Imaging parameters of vascularisation correlated with the results of histology. DWI was able to characterise both the extent of necrosis and the level of perfusion.

L’objectif de ce travail de thèse était de proposer une approche systématique et raisonnée des outils de quantification disponibles en routine clinique et du domaine de la recherche dans la prise en charge des patients atteints d’une tumeur endocrine. Nos résultats font état des limites des paramètres semi-quantitatifs pour faire le diagnostic de malignité d’un ITf fixant au 18FDG-TEP/TDM. Nos données sur l’intérêt des paramètres de textures dans le diagnostic de malignité des tumeurs surrénaliennes nous montrent des performances excellentes de ces indices mais qui n’apparaissent cependant pas supérieures en comparaison des paramètres semi-quantitatifs conventionnels. Notre approche soulève de nouvelles hypothèses quant à l’intérêt de l’analyse de texture dans la caractérisation de certaines tumeurs surrénaliennes rares (phéochromocytomes et corticosurrenalomes) qui devront faire l’objet de futurs travaux. Enfin, notre travail sur l’approche multi-paramétrique en imagerie 68Ga-DOTATOC-TEP/TDM met en évidence la faisabilité d’une acquisition dynamique corps entier en 68Ga-DOTATOC-TEP/TDM et les possibilités de quantification de la fixation tumorale des TNE traitées par RIV sur les acquisitions TEMP/TDM post-thérapeutiques. Nos résultats ouvrent des perspectives multiples quant à la prise en charge des TNE à différents étapes de la maladie (diagnostique, pronostique et évaluation thérapeutique)
The objective of this work was to propose a systematic and reasoned approach of the quantification tools available in clinical routine and in research in the management of patients with endocrine tumors. Our results report the limitations of semi-quantitative parameters to assess the malignancy of fTI detected by 18FDG-PET / CT. Our data on the interest of textural parameters in the diagnosis of malignancy of adrenal tumors show excellent performances of these indices. However these indices do not appear superior in comparison with conventional semi-quantitative parameters.Our approach raises new hypotheses regarding the interest of texture analysis in the characterization of some rare adrenal tumors (pheochromocytomas and adrenocortical carcinoma) that will be the subject of future work. Finally, our work on the multi-parametric 68Ga-DOTATOC-PET / CT imaging highlights the feasibility of a whole-body dynamic 68Ga-DOTATOC-PET / CT acquisition and the possibilities to perform quantification of NETs uptake treated with PRRT on post-therapeutic SPECT / CT acquisitions. Our results open up multiple perspectives for the management of NETs at different stages of the disease (diagnosis, prognosis and therapeutic evaluation)

A presença da Medicina Nuclear como modalidade de obtenção de imagens médicas é um dos principais procedimentos utilizados hoje nos centros de saúde, tendo como grande vantagem a capacidade de analisar o comportamento metabólico do paciente, traduzindo-se em diagnósticos precoces. Entretanto, sabe-se que a quantificação em Medicina Nuclear é dificultada por diversos fatores, entre os quais estão a correção de atenuação, espalhamento, algoritmos de reconstrução e modelos assumidos. Neste contexto, o principal objetivo deste projeto foi melhorar a acurácia e a precisão na análise de imagens de PET/CT via processos realísticos e bem controlados. Para esse fim, foi proposta a elaboração de uma estrutura modular, a qual está composta por um conjunto de passos consecutivamente interligados começando com a simulação de phantoms antropomórficos 3D para posteriormente gerar as projeções realísticas PET/CT usando a plataforma GATE (com simulação de Monte Carlo), em seguida é aplicada uma etapa de reconstrução de imagens 3D, na sequência as imagens são filtradas (por meio do filtro de Anscombe/Wiener para a redução de ruído Poisson caraterístico deste tipo de imagens) e, segmentadas (baseados na teoria Fuzzy Connectedness). Uma vez definida a região de interesse (ROI) foram produzidas as Curvas de Atividade de Entrada e Resultante requeridas no processo de análise da dinâmica de compartimentos com o qual foi obtida a quantificação do metabolismo do órgão ou estrutura de estudo. Finalmente, de uma maneira semelhante imagens PET/CT reais fornecidas pelo Instituto do Coração (InCor) do Hospital das Clínicas da Faculdade de Medicina da Universidade de São Paulo (HC-FMUSP) foram analisadas. Portanto, concluiu-se que a etapa de filtragem tridimensional usando o filtro Anscombe/Wiener foi relevante e de alto impacto no processo de quantificação metabólica e em outras etapas importantes do projeto em geral.
The presence of Nuclear Medicine as a medical imaging modality is one of the main procedures utilized nowadays in medical centers, and the great advantage of that procedure is its capacity to analyze the metabolic behavior of the patient, resulting in early diagnoses. However, the quantification in Nuclear Medicine is known to be complicated by many factors, such as degradations due to attenuation, scattering, reconstruction algorithms and assumed models. In this context, the goal of this project is to improve the accuracy and the precision of quantification in PET/CT images by means of realistic and well-controlled processes. For this purpose, we proposed to develop a framework, which consists in a set of consecutively interlinked steps that is initiated with the simulation of 3D anthropomorphic phantoms. These phantoms were used to generate realistic PET/CT projections by applying the GATE platform (with Monte Carlo simulation). Then a 3D image reconstruction was executed, followed by a filtering process (using the Anscombe/Wiener filter to reduce Poisson noise characteristic of this type of images) and, a segmentation process (based on the Fuzzy Connectedness theory). After defining the region of interest (ROI), input activity and output response curves are required for the compartment analysis in order to obtain the Metabolic Quantification of the selected organ or structure. Finally, in the same manner real images provided from the Heart Institute (InCor) of Hospital das Clínicas, Faculty of Medicine, University of São Paulo (HC-FMUSP) were analysed. Therefore, it is concluded that the three-dimensional filtering step using the Ascombe/Wiener filter was preponderant and had a high impact on the metabolic quantification process and on other important stages of the whole project.

Les maladies bronchiques obstructives chroniques constituent un problème majeur de santé publique. Leur pathogénie est caractérisée par l’inflammation chronique et le remodelage des voies aériennes. L’étude en imagerie des différents éléments structuraux des poumons (bronches, bronchioles, vaisseaux…) est primordiale dans la définition, le phénotypage, la compréhension des mécanismes obstructifs et le suivi de l’évolution et de ces maladies. Les méthodes d’évaluation visuelles qualitatives sont sujettes à une subjectivité d’analyse et à une variabilité inter observateurs qui altèrent leur fiabilité. Le développement de nouvelles méthodes quantitatives fiables et reproductibles est donc une nécessité. La TDM et l’IRM sont les deux méthodes d’imagerie complètes à la fois des différents compartiments structuraux et fonctionnels du poumon, la première ayant pour avantage son excellente résolution spatiale et la seconde son innocuité. Les travaux ont porté sur la mise au point et la validation de méthodes d’imagerie quantitatives TDM et IRM. Avec la TDM quantitative, nous avons montré que les mesures du remodelage bronchique et vasculaire chez les patients atteints d’une hypertension pulmonaire (HTP) sévère, phénotype particulier à la bronchopneumopathie chronique obstructive (BPCO), permettaient d’appréhender les interactions entre les systèmes cardiovasculaire et respiratoire. D’autre part nous avons mis au point et évalué une méthode de quantification de l’atteinte des voies aériennes distales dans une population de pneumonie chronique d’hypersensibilité. Cette méthode pourrait apporter une contribution à l’étude et au suivi des maladies bronchiques obstructives chroniques.Avec l’IRM quantitative, les solutions que nous avons proposées dans la quantification de l’emphysème à l’IRM protonique permettent une analyse automatisée de la sévérité de l’extension de l’emphysème chez des patients atteints de BPCO. En outre, la transposition des méthodes de mesure des paramètres bronchiques de la TDM à l’IRM est devenue possible grâce aux nouvelles séquences à temps d’écho ultra court (UTE). Ainsi, la quantification du remodelage bronchique à l’IRM 3DUTE chez des patients atteints de mucoviscidose, chez lesquels il est nécessaire de réduire l’exposition aux rayons ionisants, permet d’obtenir des informations morphologiques similaires à celles de la TDM. Nous avons aussi montré que la quantification automatique des phénomènes destructifs et inflammatoires par l’IRM 3DUTE dans l’atteinte pulmonaire de la mucoviscidose était une méthode fiable et reproductible pour évaluer la sévérité de l’atteinte structurale. Par ailleurs, la faisabilité d’une quantification automatique des hypersignaux T2 à l’IRM a été démontrée et sa pertinence comme biomarqueur spécifique de l’atteinte inflammatoire des voies aériennes évaluée. Ainsi, l’étude quantitative, en TDM comme en IRM, de diverses modifications structurales et fonctionnelles des maladies bronchiques obstructives chroniques ouvre le champ d’une analyse objective et fiable par l’imagerie aux évaluations du suivi et de la réponse au traitement
Chronic obstructive airway diseases are a major public health problem, characterized by chronic inflammation and airways remodeling. Imaging of the structural elements of the lungs (bronchi, bronchioles, vessels...) is essential for defining, phenotyping and following-up these diseases. Visual assessment is prone to inter-observer variability that affects its reliability. Therefore, the development of reliable and reproducible new quantitative methods is necessary. CT and MRI are the two complete imaging methods of the various structural and functional compartments of the lung. This work focused on the development and the validation of quantitative methods using CT and MRI of the lung.Using quantitative CT, we have shown that measurements of bronchial and vascular remodeling in patients with severe pulmonary hypertension (PH), a particular phenotype in COPD patients, contributed to highlight interactions between the cardiovascular and respiratory systems. We have developed a method for small airway disease quantification, tested in a chronic hypersensitivity pneumonitis cohort, that may contribute to evaluate and monitor chronic obstructive airway diseases.Using quantitative MRI, we have developed a fully automated quantification technique to assess the severity of emphysema extent in COPD patients. In addition, transposing bronchial measurement methods from CT to MRI has become possible thanks to the new ultra short echo time (UTE) sequences. Thus, the quantification of bronchial remodeling at 3DUTE MRI in patients with cystic fibrosis, for whom it is necessary to reduce exposure to ionizing radiation, has shown morphological information similar to that of CT. We have also shown that automatic quantification of destructive and inflammatory phenomena by 3DUTE MRI in cystic fibrosis is a reliable and reproducible method for assessing the severity of structural alterations. Furthermore, the feasibility of an automatic quantification of T2 high signal intensity on MRI has been demonstrated and its relevance as a specific biomarker for inflammatory airway disease has been assessed.Thus, the quantitative analysis, in both CT and MRI, of various structural and functional modifications in chronic obstructive airway diseases could be a reliable method in the follow-up and the evaluation of the response to treatment in these diseases

La Tomographie par Emission de Positons (TEP) au Fluoro-déoxyglucose marqué au Fluor 18 (¹⁸F-FDG), analogue du glucose, permet d'obtenir une image de la consommation de glucose dans l'organisme. La plupart des foyers tumoraux présentant une consommation excessive de glucose, son utilisation en oncologie permet d'améliorer la prise en charge des patients en diminuant le temps nécessaire pour évaluer l'efficacité des traitements tels que la chimiothérapie et la radiothérapie. Mon projet de recherche visait à proposer et améliorer des méthodes de quantification en TEP au ¹⁸F-FDG afin de caractériser au mieux l'évolution métabolique des volumes tumoraux.De nombreux facteurs biaisent la quantification en TEP. Parmi eux, l'Effet de Volume Partiel (EVP) reste difficile à corriger, notamment à cause de la faible résolution spatiale des images TEP. Afin de déterminer l'impact de la correction de l'EVP sur l'évaluation des réponses des tumeurs, une étude sur données simulées par Monte Carlo a tout d'abord été effectuée. Cette étude a été complétée par l'analyse de données TEP/TDM (Tomodensitométrie) acquises chez 40 patients atteints de cancers colorectaux métastatiques (CCM), traités par chimiothérapie à l'Institut Jules Bordet (Bruxelles). L'analyse de 101 tumeurs a montré que les critères tels que le SUV, n'incluant pas de correction de l'EVP, et qui reflètent alors le volume tumoral et son activité, prédisaient mieux l'évolution tumorale que les critères corrigés de l'EVP. Compte tenus des résultats prometteurs récents de méthodes de caractérisation de l'hétérogénéité de la fixation du FDG dans les tumeurs, un second volet de notre travail a consisté à étudier l'intérêt de la prise en compte de la texture dans le cadre du suivi thérapeutique. L'application de l'analyse de texture aux cas de CCM étudiés précédemment n'a pas permis de démontrer une valeur ajoutée des indices de texture par rapport aux index quantitatifs couramment employés en clinique. Nous avons montré que cette conclusion s'expliquait en partie par la non-robustesse des indices de texture vis-à-vis des paramètres impliqués dans leur mesure. Nous avons enfin cherché à évaluer une méthode d'Analyse Factorielle de Séquences d'Images Médicales (AFSIM), appliquée au contexte du suivi thérapeutique, pour caractériser l'évolution tumorale tout au long du traitement. Cette étude a porté sur 9 séries de 4 à 6 examens TEP/TDM de patients traités par radiothérapie au Centre Hospitalier Universitaire Henri Becquerel de Rouen. Outre l'information visuelle globale apportée par cette méthode, l'analyse quantitative des résultats obtenus a permis de caractériser l'hétérogénéité de la réponse vue par l'AFSIM. L'échec des index classiques, provenant entre autres de leur incapacité à distinguer les processus inflammatoires de l'activité métabolique tumorale, a permis de monter la valeur ajoutée de l'AFSIM par rapport aux index tels que le SUV maximal ou moyen.

Im Rahmen der vorliegenden Dissertationsschrift „Semi-quantitative röntgentomographische Untersuchungen zur Biodistribution von magnetischen Nanopartikeln in biologischem Gewebe“ wurden tomographische Untersuchungen an biologischen Objekten durchgeführt. Bei diesen Objekten handelt es sich um Gewebeproben nach minimal-invasiven Krebstherapien wie zum Beispiel magnetischem Drug Targeting und magnetischer Wärmebehandlung. Der Erfolg dieser Therapien ist sowohl abhängig von der korrekten Verteilung der magnetischen Nanopartikel als auch von der Tatsache, dass diese in der Zielregion in einer ausreichenden Menge vorhanden sind. Das Vorliegen dieser beiden Voraussetzungen ist in der vorliegenden Arbeit untersucht worden. Dabei lag der Schwerpunkt der Arbeit auf der Quantifizierung von magnetischem Material in unterschiedlichen biologischen Gewebeproben mittels Röntgenmikrocomputertomographie (XµCT). Für diesen Zweck wurde ein Kalibrationssystem mit speziellen Phantomen entwickelt, mit dessen Hilfe eine Nanopartikelkonzentration einem Grauwert voxelweise zugewiesen werden kann. Mit Hilfe der Kalibration kann der Nanopartikelgehalt sowohl in monochromatischen als auch in polychromatischen tomographischen Daten im Vergleich zu magnetorelaxometrischen Ergebnissen mit wenigen Prozent Abweichung ermittelt werden. Trotz Polychromasie und damit einhergehenden Artefakten können 3-dimensionale röntgentomographische Datensätze mit einer geringfügigen Konzentrationsabweichung im Vergleich zur quantitativen Messmethode Magnetorelaxometrie semi-quantitativ ausgewertet werden
The success of the minimal invasive cancer therapies, called magnetic drug targeting and magnetic heating treatment, depends strongly on the correct distribution of the magnetic nanoparticles on one side. On the other side it depends on the fact that a sufficient amount of magnetic nanoparticles carrying drugs is accumulated in the target region. To study whether these two requirements are fulfilled motivates this PhD thesis „Semi-quantitative X-ray-tomography examinations of biodistribution of magnetic nanoparticles in biological tissues“. The analysis of the distribution of the magnetic nanoparticles in tumours and other tissue examples is realized by means of X-ray-micro computer tomography (XμCT). The work focuses on the quantification of the magnetic nanoparticles in different biological tissue samples by means of XµCT. A calibration of the tomographic devices with adequate phantoms, developed in the frame of this work, opens now the possibility to analyze tomographic data in a semi-quantitative manner. Thus, the nanoparticle concentration can be allocated voxel-wise to the grey values of the three-dimensional tomographic data. With the help of calibration of the tomography equipments used, polychromatic as well as monochromatic three-dimensional representations of objects can be analyzed with regard to the biodistribution of magnetic nanoparticles as well as with regard to their quantity. The semi-quantitative results have been compared with results obtained with a quantitative measurement method magnetorelaxometry (MRX). Thereby a good agreement of the semi-quantitative and quantitative data has been figured out

碩士
中原大學
醫學工程研究所
90
In clinical, high-resolution chest computed tomography(CT) image can supports three-dimensional(3D) viewpoint and understanding in anatomy. According to the image texture, it could give the basics of clinical decision and the directions. Thus it could deal with pulmonary diseases and the changes of disease’s transfer. In this thesis, we developed a system which could acquire the pulmonary function parameters from chest spiral CT image sets and helped physician using those to make decisions in clinical diagnosis. The system read serial chest CT DICOM image sets directly for analysis and used different threshold to process different thickness of the image slice that could separate the lung area from that. The semi-automatic image segmentation technology was used for this system were adopted masks to reduce the noise and image processing functions were used to analysis image and find the pulmonary function parameters. The image processing functions included Region-grow, Ray-casting, and Voxel-highlighting technology. The pulmonary function parameters included lung capacity, the thickness of airway tree, the lumen of airway tree, bullae index (BI), and the angle and direction of airway tree. Besides, this system could show 3D surface render image which composed of the OpenGL API and Marching cube algorithm, was needed in anatomy. Finally, this system could analysis phantom image, normal and patient image. The accuracy and the applicability of system parameter also were evaluated. By the result, the accuracy of system could be evaluated by using the phantom image. Applying this system to analysis normal and patient image could get following results. Firstly, the R.R value in lung capacity was 0.910 between the traditional lung function test and system calculated, was low as a result of the different of the pose in measure and the effort of partial volume. Secondly, the p-value in t-test of the “airway lumen /airway thickness” parameter which compared by the normal and Chronic Obstructive Pulmonary Disease (COPD) patient was lower than 0.05. Finally, the BI index could present the class 3 and class 4 parts were important for detecting the degree of emphysema. Because the p-value was lower than 0.05, it showed number of point and percent of class 3 and class 4 could distinguish between normal and patient also provided the quantification in the degree of emphysema. This research could provide a system to complete the 3D object reconstructed and pulmonary function parameters acquired which could be applied the clinical application to diagnose the lung disease. In the further, the system might combine with the database, and then using the data mining and other analysis technology assist the clinical applications.

The work presented in this dissertation focuses on evaluation of absolute quantification accuracy in dedicated breast SPECT-CT. The overall goal was to investigate through simulations and measurements the impact and utilization of various correction methods for scattered and attenuated photons, characterization of incomplete charge collection in Cadmium Zinc Telluride detectors as a surrogate means of improving scatter correction, and resolution recovery methods for modeling collimator blur during image reconstruction. The quantification accuracy of attenuation coefficients in CT reconstructions was evaluated in geometric phantoms, and a slice-by-slice breast segmentation algorithm was developed to separate adipose and glandular tissue. All correction and segmentation methods were then applied to a pilot study imaging parathyroid patients to determine the average uptake of Tc-99m Sestamibi in healthy breast tissue, including tissue specific uptake in adipose and glandular tissue.

Monte Carlo methods were utilized to examine the changes in incident scatter energy distribution on the SPECT detector as a function of 3D detector position about a pendant breast geometry. A simulated prone breast geometry with torso, heart, and liver was designed. An ideal detector was positioned at various azimuthal and tilted positions to mimic the capabilities of the breast SPECT subsystem. The limited near-photopeak scatter energy range in simulated spectra was linearly fit and the slope used to characterize changes in scatter distribution as a function of detector position. Results show that the detected scatter distribution changes with detector tilt, with increasing incidence of high energy scattered photons at larger detector tilts. However, reconstructions of various simulated trajectories show minimal impact on quantification (<5%) compared to a primary-only reconstruction.

Two scatter compensation methods were investigated and compared to a narrow photopeak-only windowing for quantification accuracy in large uniform regions and small, regional uptake areas: 1) a narrow ±4% photopeak energy window to minimize scatter in the photopeak window, 2) the previously calibrated dual-energy window scatter correction method, and 3) a modified dual-energy window correction method that attempts to account for the effects of incomplete charge collection in Cadmium Zinc Telluride detectors. Various cylindrical phantoms, including those with imbedded hot and cold regions, were evaluated. Results show that the Photopeak-only and DEW methods yield reasonable quantification accuracy (within 10%) for a wide range of activity concentrations and phantom configurations. The mDEW demonstrated highly accurate quantification measurements in large, uniform regions with improved uniformity compared to the DEW method. However, the mDEW method is susceptible to the calibration parameters and the activity concentration of the scanned phantom. The sensitivity of the mDEW to these factors makes it a poor choice for robust quantification applications. Thus, the DEW method using a high-performance CZT gamma camera is still a better choice for quantification purposes

Phantoms studies were performed to investigate the application of SPECT vs CT attenuation correction. Minor differences were observed between SPECT and CT maps when assuming a uniformly filled phantom with the attenuation coefficient of water, except when the SPECT attenuation map volume was significantly larger than the CT volume. Material specific attenuation coefficients reduce the corresponding measured activity concentrations compared to a water-only correction, but the results do not appear more accurate than a water-only attenuation map. Investigations on the impact of image registration show that accurate registration is necessary for absolute quantification, with errors up to 14% observed for 1.5cm shifts.

A method of modeling collimator resolution within the SPECT reconstruction algorithm was investigated for its impact on contrast and quantification accuracy. Three levels of resolution modeling, each with increasing ray-sampling, were investigated. The resolution model was applied to both cylindrical and anthropomorphic breast phantoms with hot and cold regions. Large volume quantification results (background measurements) are unaffected by the application of resolution modeling. For smaller chambers and simulated lesions, contrast generally increases with resolution modeling. Edges of lesions also appear sharper with resolution modeling. No significant differences were seen between the various levels of resolution modeling. However, Gibbs artifacts are amplified at the boundaries of high contrast regions, which can significantly affect absolute quantification measurements. Convergence with resolution modeling is also notably slower, requiring more iterations with OSEM to reach a stable mean activity concentration. Additionally, reconstructions require far more computing time with resolution modeling due to the increase in number of sampling rays. Thus while the edge enhancement and contrast improvements may benefit lesion detection, the artifacts, slower convergence, and increased reconstruction time limit the utility of resolution modeling for both absolute quantification and clinical imaging studies.

Finally, a clinical pilot study was initiated to measure the average uptake of Tc-99m Sestamibi in healthy breast tissue. Subjects were consented from those undergoing diagnostic parathyroid studies at Duke. Each subject was injected with 25mCi of Sestamibi as part of their pre-surgical parathyroid SPECT imaging studies and scanned with the dedicated breast SPECT-CT system before their diagnostic parathyroid SPECT scan. Based on phantom studies of CT reconstructed attenuation coefficient accuracy, a slice-by-slice segmentation algorithm was developed to separate breast CT data into adipose and glandular tissue. SPECT data were scatter, attenuation, and decay corrected to the time of injection. Segmented CT images were used to measure average radiotracer concentration in the whole breast, as well as adipose and glandular tissue. With 8 subjects scanned, the average measured whole breast activity concentration was found to be 0.10µCi/mL. No significant differences were seen between adipose and glandular tissue uptake.

In conclusion, the application of various characterization and correct methods for quantitative SPECT imaging were investigated. Changes in detected scatter distribution appear to have minimal impact on quantification, and characterization of low-energy tailing for a modified scatter subtraction method yields inferior overall quantification results. Comparable quantification accuracy is seen with SPECT and CT-based attenuation maps, assuming the SPECT-based volume is fairly accurate. In general, resolution recovery within OSEM yields higher contrast, but quantification accuracy appears more susceptible to measurement location. Finally, scatter, attenuation, and resolution recovery methods, along with a breast segmentation algorithm, were implemented in a clinical imaging study for quantifying Tc-99m Sestamibi uptake. While the average whole breast uptake was measured to be 0. 10µCi/mL, no significant differences were seen between adipose and glandular tissue or when implementing resolution recovery. Thus, for future clinical imaging, it's recommended that the application of the investigated correction methods should be limited to the traditional DEW method and CT-based attenuation maps for quantification studies.

Dissertation

碩士
國立陽明大學
生物醫學影像暨放射科學系
107
Background and Purpose: Positron emission tomography/magnetic resonance imaging (PET/MR) combines two imaging modalities to produce hybrid images with both advantageous qualities. Hybrid images from PET/MR present exquisite anatomical and structural details provided by MRI, and functional and molecular imaging from PET. Several difficulties have to be overcome in order to achieve an accurate quantification of PET, this includes physical effects, such as attenuation, scatter, random coincidences, and detector efficiency normalization. Attenuation correction (AC) is vital in PET, as attenuation of high energy photons by tissues and materials may lead to underestimation or overestimation of tracer uptake causing artifacts and misdiagnosis. CTAC (computed tomography based attenuation correction) is considered the golden standard when it comes to AC. CTAC involves accounting photon attenuation coefficients along each line of response (LOR), μ (attenuation) maps can thus be derived with the coefficients acquired, further serving as a template for AC of the PET images. However in PET/MR imaging, unlike values in computed tomography (CT) which provide electron density and photon attenuation properties of tissues, MRI intensities reflect proton density and relaxation time properties. The values from MRI don’t entirely correlate to tissue density and cannot be converted directly to μ maps, therefore MRAC (magnetic resonance based attenuation correction) is not considered accurate. The SUV (standard uptake value) and radioactivity (Bq/ ml), which are acquired from PET/MR for quantification are also affected by MRAC. Methods and Materials: In our study, various phantoms were scanned with PET/CT and PET/MR scanners, and attenuation corrected with the scanner’s respective methods. CTAC was done with CT based μ-maps, and MRAC was done through Dixon derived μ-maps, the SUV and radioactivity (Bq/ ml) results from AC of PET/MRI and PET/CT were further compared by circling ROIs (region of interest) and methods to minimize their differences were developed. Conclusion: The study aims to provide methods to harmonize quantification in PET/MR and PET/CT modalities, providing means to compare results acquired from the two. The results revealed that radioactivity and SUVs between the two scanners had high correlation and can be converted with linear regression formulas, for abdomen region SUVPETCT = 1.3103 × SUVPETMR + 0.0541, for lung region SUVPETCT = 1.762 × SUVPETMR + 0.0929, and for head region SUVPETCT = 1.5688 × SUVPETMR + 0.2821. Through harmonization of the two modalities, we are a step closer to achieving “precision diagnostics”.

This work investigates phantom and subject positioning as well as collecting data with a variety of angular sampling and acquisition trajectories. The overall goal of this work has been to utilize the dedicated, breast SPECT-CT system to acquire the best possible images. A large portion of this work has been to apply corrections to the system for quantitative imaging. The system has been shown to provide high quality images with minimal out-of-field signal contribution. Additionally, the quantification procedure has been shown to be within 10% of the known activity concentration present at the time of imaging for both VAOR and PROJSINE trajectories.

Dissertation

Introduction: Despite the relatively low dose (0.5 mSv – 1 mSv) generated by Attenuation Correction CT (ACCT) in PET examinations, the ALARA principle is still applicable. The currently used ACCT standard protocol at the Karolinska University Hospital in Solna uses 7.6 effective mAs (mAseff) and 120 kVp, but reducing mAseff and/or kVp would decrease patient dose as well as facilitate an increased number of research subjects. A CT reconstruction algorithm called Quantification Achieved Consistently (Q.AC.) (Lonn, 2012) has recently been developed to enable reduced doses from ACCT, while preserving quantitative PET data. The purposes of this study were to investigate possible limitations of the Q.AC. with respect to patient size, and to optimise protocols, aiming at minimising ACCT dose in terms of Volumetric Computer Tomography Dose Index CTDIvol. Methods: Measurements were performed with a GE PET/CT Discovery system, which offers Q.AC. reconstruction. The NEMA NU-2 protocol was followed to quantify PET quality, including evaluations of relative count error in the artificial lung in the phantom centre (lung), hot- and cold-sphere contrast (Q), and background variability (N). Two phantoms were used; the NEMA body phantom (elliptical cross section sized 30 cm laterally and 23 cm anterior-posterior (AP)), here representing paediatric patients and small-sized adults, and the same phantom with an additional (20 cm laterally and 4 cm AP) ellipsoid plastic (PMMA) extension ring, representing mid- and large-sized patients. ACCTs were acquired with 15 mAseff values, range [2.3 - 260], in combination with four kVp values [80, 100, 120, 140] and reconstructed with two algorithms (Q.AC. and a regular soft CT algorithm). Consequently, PET reconstructions were performed based on each mAseff, kVp and CT-reconstruction combination. Results: Quantitatively similar PET results to the standard protocol were achieved with the Q.AC. CT reconstruction algorithm using a CTDIvol = 0.06 mGy (2.3 mAseff and 80 kVp) for the NEMA body phantom, respectively a CTDIvol = 0.20 mGy (2.3 mAseff and 120 kVp) for the phantom with additional extension ring. Conclusions: This study indicates that the Q.AC. CT reconstruction algorithm enables accurate PET results at lower ACCT mAseff and kVp settings than the currently used clinical standard protocol. For paediatric patients and small-sized adults, a reduction of CTDIvol by approximately 90% may be achieved, while for mid- and large-sized patients, the CTDIvol can be reduced by approximately 70% without loss of quantitative PET data.

Inflammation in response to infection and injury is a critical survival mechanism used by all higher vertebrates. Chronic inflammatory conditions are associated with the prolongerelease of inflammatory mediators and the activation of harmful signal transduction pathways, all of which contribute disease development and phenotypes. Extracellular Volume Fraction (ECV) may be able to detect subtle abnormalities such as diffuse inflammation acute or chronic due to infection and/or fibrosis. The validity of this technique was preliminarily evaluated in a study with 20 patients suspected to have diffuse inflammation in the myocardial tissue. Using HU (Hounsfield Unit) values before and after administration of an Extracellular Contrast Agent (ECA) allows the additional calculation of the ECV, well established in CT. In fact, the ratio of the change in blood and tissue attenuation (HU) represents the contrast agent partition coefficient. People living with HIV (PLWH) have an increase vulnerability to sub-clinical and clinical cardiovascular (CV) diseases. Purpose: PLWH are prone to develop sub-clinical Cardiovascular (sCV) disease, despite the effectiveness of combined AntiRetroviral Therapy (cART). Algorithms developed to predict CV risk in the general population could be inaccurate when applied to PLWH. Myocardial Extra-Cellular Matrix (ECM) expansion, measured by computed tomography, has been associated with an increased CV vulnerability in HIV-negative population. Measurement of Myocardial ECV by computed tomography or magnetic resonance, is considered a useful surrogate for clinical evaluation of ECM expansion. In the present study, we aimed to determine the extent of cardiovascular involvement in asymptomatic HIV-infected patients with the use of a comprehensive cardiac computed tomography (CCT) approach. In this study, ECV in low atherosclerotic CV risk PLWH was compared with ECV of age and gender matched HIV- individuals. 53 asymptomatic HIV+ individuals (45 males, median age 48 (IQR:42.5-48) years) on effective cART (CD4+ cell count: 450 cells/μL (IQR: 328-750); plasma HIV RNA: <37 copies/ml in all subjects) and 18 age and gender matched controls (14 males, median age 55 (IQR:44.5-56) years) were retrospectively enrolled. All participants underwent CCT protocol to obtain native and postcontrast Hounsfield unit values of blood and myocardium, ECM was calculated accordingly. The ECV was significantly higher in HIV+ patients than in the control group (ECV: 31% (IQR: 28%-31%) vs 27.4% (IQR: 25%-28%), p<0.001). The duration of cART (standardized=0.56 (0.33-0.95), p=0.014) and the years of exposure to HIV infection [standardized=0.53 (0.4-0.92), p<0.001], were positively and strongly associated with ECV values. Differences in ECV (p<0.001) were also observed regarding the duration of exposure to cART (<5 years, 5-10 years and >10 years). Moreover, ECV was independently associated with age of participants [standardized = 0.42 (0.33-0.89), p=0.084]. We hypothesized that quantitative assessment of tissue ECV would be clinically useful for detecting both focal and diffuse tissue abnormalities in a variety of acute and chronic infectious conditions. ECV imaging can quantitatively characterize tissue inflammation, atypical diffuse fibrosis, and subtle tissue abnormalities not clinically apparent on different method images. Therefore, ECV not only can detect tissue inflammation and/or fibrosis but also might quantify response to treatment during follow-up. HIV infection and exposure to antiretrovirals play a detrimental role on ECV expansion. An increase in ECV indicates ECM expansion, which has been associated to a higher CV risk in the general population. The non-invasive evaluation of ECM trough ECV could represent an important tool to further understand the relationship between HIV infection, cardiac pathophysiology and the increased CV risk observed in PLWH.

博士
國立陽明大學
生物醫學影像暨放射科學系
100
Perfusion is the process of nutritive delivery of arterial blood to a capillary bed in the biological tissues. Sufficient tissue perfusion and oxygenation are vital for all metabolic processes in cells. Over the past few years, stroke has been the third leading cause of death in Taiwan and the rest of the world. Stroke occurs when the blood flow to the brain is impaired, resulting in an insufficient supply of nutrients and oxygen. Carotid stenosis is one of the many causes of stroke. Several imaging techniques have been applied to assess cerebral perfusion, such as positron emission tomography (PET), computed tomography (CT), and dynamic susceptibility contrast magnetic resonance imaging (DSC–MRI). Some reports have compared perfusion studies of computed tomography perfusion (CTP) and PET in normal volunteers and in patients with chronic cervical carotid artery occlusion. Quantitative imaging is a key factor for the diagnosis in the clinical examination; however, absolute quantification in DSC–MRI has not been fully investigated. The aim of the study was to assess absolute quantification of dynamic susceptibility contrast–enhanced magnetic resonance perfusion (MRP) comparing with computed tomography perfusion (CTP) as the standard reference in patients with unilateral stenosis. We retrospectively post-processed MRP in 20 patients with unilateral occlusion or stenosis of > 79% at the internal carotid artery (ICA) or the middle cerebral artery (MCA). Absolute quantification of MRP was performed after applying the following techniques: cerebral spinal fluid (CSF) removal, vessel removal using both relative cerebral blood volume (rCBV) and relative cerebral blood flow (rCBF) masks. A thresholding technique was applied to rCBV and rCBF images to remove the effects of vessel voxels. The thresholds were set at twice of the median values of the rCBV and rCBF histograms. Voxels with rCBV or rCBF values higher than the thresholds were classified as vessel voxels. The vessel voxels were removed from the brain mask. The thresholding technique was also applied to TTP images for finding normal brain parenchyma. The averaged rCBV and rCBF for the voxels segmented as normal brain parenchyma voxels were calculated for finding scaling factors to convert the rCBV and rCBF images into absolute values. The correlation between MRP and CTP was best for mean transit time (MTT) (r = 0.83), followed by time to peak (TTP) (r = 0.50), cerebral blood flow (CBF) (r = 0.52) and cerebral blood volume (CBV) (r = 0.43). There was no significant difference between MRP and CTP for CBV, CBF, MTT and TTP on the lesion side, the contralateral side, the lesion-contralateral differences, or the lesion-to-contralateral ratios (P > 0.05). The limits of agreement between MRP and CTP were as follows: CBV -0.57 ± 1.15 mL/100 g, CBF 2.50 ± 17.01 mL/100 g/min, MTT -0.90 ± 2.76 sec, and TTP 3.83 ± 7.01 sec. Absolute quantification of MRP is possible. Using the proposed method, the measured values of MRP and CTP had acceptable linear correlation and acceptable quantitative agreement.

Dissertation Abstract Abdominalaortic aneurysm (AAA) is a serious disease. Its prevalence is in the developed countries about 3%. As an aneurysm is considered a dilatation of all layers of a vessel wall over 3 cm. Majority of AAA are small and asymptomatic, and although the risk of rupture increases with the size of aneurysm sack, even the small aneurysms rupture. The rupture mortaliry ranges about 70 %. Surgical treatment is indicated in the asymptomatic patients in diameter of AAA over 5 cm. The average speed of growth of AAA is 0.3 cm per year, e.g. in the early diagnosed patients there is a several years interval for a pharmacolocical influencing of the progression of this disease. Knowledge of pathogenesis is essential for any targeted pharmacological treatment. Our prospective, non-randomised studies are based on the application of the stereological methods for the histopathological assessment of the AAA samples. The acquired data enable the statistical analysis, including the null hypothesis testing. In our study analyzing the histopathology of AAA aortae of 65 patients (65 walls and 55 thrombi) and 6 normal abdominal aortae from the organ donors we assessed the following parameters: the area fractions of collagen and elastin, and the length density of elastin in intima and media, the area...

Tese de doutoramento em Ciências da Saúde, no ramo de Medicina, na especialidade de Medicina Interna (Medicina Nuclear), apresentada à Faculdade de Medicina da Universidade de Coimbra
A PET/CT possibilita a obtenção de imagens médicas em que se encontram conjugadas informações de tipo estrutural (CT) com informações de natureza molecular (PET). A sua importância clínica tem apresentado um crescimento exponencial sendo hoje uma técnica fundamental para a avaliação de múltiplas situações do foro oncológico. Esse crescimento encontra-se justificado pelo desenvolvimento constantemente verificado nos equipamentos utilizados, bem como na disponibilização progressiva de novos radiofármacos. Apesar da disponibilidade crescente de outros radionuclídeos emissores de positrões, o flúor-18 continua a ser o mais escolhido para a marcação de radiofármacos PET. Esta preferência deve-se ao facto de o flúor-18 reunir um conjunto vantajoso de características físicas, químicas, radioquímicas e de disponibilidade. Dos vários radiofármacos PET usados em oncologia, o mais utilizado é a [18F]FDG, correspondendo a cerca de 90% dos estudos PET/CT realizados. No entanto, outros radiofármacos marcados com flúor-18, de introdução mais recente, têm vindo a adquirir importância clínica significativa. Esta Tese pretende ser um contributo para a valorização do interesse clínico da utilização de alguns radiofármacos PET marcados com flúor-18. São apresentados os resultados obtidos em diversas situações oncológicas através da utilização de [18F]FDG bem como de outras moléculas, nomeadamente [18F]NaF, [18F]FCH e [18F]FMISO. São também avaliadas novas abordagens de interpretação dos exames efetuados, em que se valoriza o conceito radiomics, ou seja a quantificação de parâmetros extraídos das imagens obtidas. Em doentes com carcinoma de não pequenas células do pulmão avaliados por PET/CT com [18F]FDG ficou demonstrado o valor prognóstico da quantificação da carga tumoral total metabolicamente ativa, relacionando-se com a sobrevivência global dos doentes. Esta constatação sugere a existência de vantagens na conjugação desta informação com o estadiamento cTNM para efeito de avaliação prognóstica e de decisão terapêutica. Também em doentes com carcinoma de não pequenas células do pulmão se demonstrou que a avaliação radiómica acresce informação à habitual interpretação da PET/CT com [18F]FDG. As informações quantitativas obtidas relacionam-se com a sobrevivência global dos doentes e apresentam elevado valor prognóstico. Nestes doentes foi, também, testado um novo software para identificação e quantificação de heterogeneidade tumoral, característica que se relaciona com maior probabilidade de recidiva e com má resposta à terapêutica. Ficou comprovada a superioridade da avaliação quantitativa de heterogeneidade tumoral quando comparada com a análise visual das imagens. Em doentes oncológicos com suspeita de metastização óssea foi testada a possibilidade da execução de PET/CT através de uma utilização combinada de [18F]FDG e [18F]NaF. O estudo prospetivo e multicêntrico efetuado confirmou a exequibilidade dessa metodologia e aponta as suas potenciais vantagens na avaliação global de metastização óssea e extraóssea. Ficou demonstrada a superior capacidade da PET/CT com [18F]NaF para a deteção de metástases ósseas quando comparada com a cintigrafia óssea, a PET/CT com [18F]FDG e a PET/CT com [18F]FCH. Foi igualmente demonstrado que a quantificação do parâmetro maximum standardized uptake value (SUVmax) pode acrescentar informação à análise visual, contribuindo para a diferenciação entre lesões degenerativas e metastáticas, e entre metástases de natureza lítica e blástica. Foi avaliada uma nova metodologia de quantificação de carga tumoral óssea em doentes com metastização óssea, aplicando-a a uma patologia ainda não avaliada desta forma (carcinoma da mama) para quantificação da resposta da doença óssea à terapêutica realizada. Em doentes com carcinoma da próstata confirmou-se o valor da PET/CT com [18F]FCH na sua abordagem diagnóstica, quer em doentes em recidiva bioquímica quer em doentes em fase de estadiamento inicial. Evidenciou-se, também, o seu contributo na escolha da opção terapêutica e no planeamento da radioterapia. Os dois únicos exames PET/CT com [18F]FMISO realizados, até agora, em Portugal, são apresentados, sendo sublinhada a importância desta técnica na identificação de subvolumes tumorais hipóxicos para a otimização do planeamento da radioterapia com incrementos de dose nesses subvolumes radiorresistentes. Com o recurso à PET/CT tem sido possível uma abordagem cada vez mais personalizada dos cuidados médicos, ao permitir, para cada doente, uma definição mais correta da opção terapêutica bem como ao informar sobre a duração ideal do seu tratamento. O propósito desta Tese é, de algum modo, contribuir para a continuação deste percurso.
PET/CT provides images in which structural information (CT) is combined with molecular information (PET). Its clinical importance has increased exponentially and it is now a fundamental technique for the evaluation of multiple oncologic situations. This increased importance is due to the constant development of the equipment used, as well as the progressive availability of new radiopharmaceuticals. Despite the increasing availability of other positron-emitting radionuclides, fluorine-18 continues to be the most popular choice for the labeling of PET radiopharmaceuticals. This preference is due to the fact that fluorine-18 possesses an advantageous set of physical, chemical, radio-chemical and availability characteristics. Of the various PET radiopharmaceuticals used in oncology, the most commonly used is [18F]FDG, corresponding to about 90% of the PET/CT studies performed. However, other newer fluoride-labeled radiopharmaceuticals have also been gaining significant clinical importance. This Thesis contributes to the validation of the clinical interest in the use of some radiopharmaceuticals labeled by fluorine-18. The results obtained in various oncological situations are shown through the use of [18F]FDG as well as other molecules such as [18F]NaF, [18F]FCH, [18F]FMISO. New approaches to interpreting the exams are also evaluated, in which the concept of radiomics, i.e the extraction of quantified parameters from the images obtained, is valued. In patients with non-small cell lung cancer evaluated with [18F]FDG PET/CT, it was shown the prognostic value of the quantified metabolic active total tumor burden relating to patients’ overall survival. This finding suggests that there may be advantages in combining this information with the cTNM staging for prognostic evaluation and therapeutic decision. It was also shown that radiometric assessment adds information to the usual interpretation of [18F]FDG PET/CT in patients with non-small cell lung cancer. The quantitative information obtained was related to patients’ overall survival and had a high prognostic value. In these patients, new software was also tested in the identification and quantification of tumor heterogeneity, a characteristic that is related to a greater probability of relapse and poor response to therapy. The superiority of the quantitative evaluation of tumor heterogeneity was verified when compared with the visual analysis of the images. In cancer patients with suspected bone metastasis, the feasibility of carrying out PET/CT was tested through a combined use of [18F]FDG and [18F]NaF. A prospective and multicenter study confirmed the feasibility of this approach and highlighted its potential advantages in the overall evaluation of bone and extra-osseous metastases. The superior capacity of [18F]NaF PET/CT was demonstrated for the detection of bone metastases when compared with bone scintigraphy, [18F]FDG PET/CT, and [18F]FCH PET/CT. It was also shown that the quantification of the maximum standardized uptake value (SUVmax) parameter can add information to the visual analysis, contributing to the differentiation between degenerative and metastatic lesions, and between lytic and blastic metastases. A new method of skeletal tumor burden quantification was also tested in patients with bone metastases. The method was applied to a condition not yet evaluated in this way (breast cancer) for quantification of skeletal disease response to therapy. The value of [18F]FCH PET/CT in the diagnostic approach for patients with prostate cancer was confirmed, both for patients with biochemical relapse and patients in the initial staging phase. There was also evidence for the contribution of this method towards the choice of therapeutical option and the planning of radiotherapy. The two PET/CT studies carried out so far in Portugal with [18F]FMISO are also presented underlining the importance of this technique for the identification of hypoxic tumor subvolumes and for the optimization of radiotherapy planning with dose increments at those radio-resistant subvolumes. PET/CT has made it possible to adopt an increasingly personalized approach to medical care by allowing each patient a more accurate selection of therapeutic options as well as informing the ideal duration of treatment. The purpose of this Thesis is to contribute to the continuation of this course.

Trabalho Final do Mestrado Integrado em Medicina apresentado à Faculdade de Medicina
INTRODUÇÃO - O carcinoma do pulmão é a maior causa de mortalidade relacionada com cancro a nível mundial, sendo o carcinoma do pulmão de não pequenas células (CPNPC) o tipo histológico mais frequente. A tomografia por emissão de positrões com tomografia computarizada com 18-fluor-2-deoxi-D-glucose (PET/CT [18F]FDG) é considerada uma ferramenta fundamental para o diagnóstico, estadiamento, planeamento terapêutico e avaliação de resposta terapêutica no CPNPC. A PET/CT [18F]FDG pode ser interpretada segundo parâmetros quantitativos tais como volume metabólico tumoral (MATVwb) e glicólise total da lesão (TLGwb). O MATVwb é obtido somando os volumes de todas as lesões metabolicamente ativas na PET/CT [18F]FDG. O TLGwb resulta do produto do MATVwb com o SUVmédio, refletindo a carga tumoral total. Esta definição sugere que o TLGwb poderá ser mais sensível na predição da sobrevivência global em doentes com CPNPC.OBJETIVO - Comparar o valor prognóstico dos parâmetros MATVwb e TLGwb em doentes com CPNPC, para melhor estadiar estes doentes. MÉTODOS E RESULTADOS - Fez-se o estadiamento de 334 doentes com CPNPC, sem metástases cerebrais ou história de outras neoplasias, entre janeiro de 2011 e agosto de 2018, de acordo com a oitava edição do estadiamento TNM para CPNPC. O estudo incluiu 92 (27.5%) mulheres e 242 (72.5%) homens, com idades entre os 33 e os 38 anos (média 66,16 ± 10,21 anos), que foram avaliados retrospetivamente. O MATVwb e o TLGwb foram quantificados, com recurso à PET/CT [18F]FDG. O tempo de sobrevivência foi analisado pelo método de Kaplan-Meier, usando o teste do Log-Rank quando apropriado, tendo a regressão de Cox sido aplicada para identificar preditores daquele e avaliado o C Index dos marcadores de prognóstico. Os testes foram avaliados ao nível de significância de 5%. Foram calculados intervalos de confiança a 95% após bootstrapping. Os doentes foram seguidos entre 0,50 e 98,3 meses (média 42,40 + 2,27 meses e mediana 22,63 + 2,61). A sobrevivência global ao ano e aos cinco anos foi, respetivamente, 97,0% e 85,4%, com uma taxa de mortalidade global de 84,6%.Os resultados revelaram que quer o MATVwb quer o TLGwb, ajustados a idade do doente, conjuntamente com o estadiamento TNM, têm um valor prognóstico elevado, e que o valor prognóstico do TLGwb nunca é inferior ao do MATVwb. CONCLUSÃO - O MATVwb e o TLGwb são ambos fortes preditores de sobrevivência global nos doentes com CPNPC. Sendo que o TLGwb, por definição, faz mais sentido em termos clínicos, prova-se que o seu valor prognóstico não é inferior ao do MATVwb.
PURPOSE - To compare the prognostic value of the parameters metabolic active tumor volume (MATVwb) and total lesion glycolysis (TLGwb) in non-small cell lung cancer (NSCLC) patients, to allow their further staging stratification. METHODS - Initial TNM staging of 334 NSCLC patients was performed, in patients without brain metastases or history of other malignancies, recruited between January/2011 and August/2018, staged according to the eighth edition of TNM staging system for NSCLC, in similarity with the previous article by Lapa et al. [1]. The study included 92 (27.5%) women and 242 (72.5%) men, aged between 33 and 88 years (66.16 ± 10.21), who were retrospectively evaluated. MATVwb and TLGwb were quantified using [18F]FDG PET/CT. Survival time was analyzed through the Kaplan-Meier method, applying the log-rank test whenever appropriate, and the Cox regression was applied in order to identify predictors of overall survival. The C Index was obtained for identified predictors. Statistical tests were evaluated at a 5% significance level and 95% confidence intervals were obtained after bootstrapping. RESULTS - Patients’ follow-up time ranged between 0.50 and 98.3 months (mean 42.40 + 2.27, median 22.63 + 2.61 months). The one and five-years survival rate were 97.0% and 85.4%, with an overall mortality rate of 84.6%.The results revealed that either MATVwb or TLGwb, adjusted for patients age, along with stage or sub-stage of the disease have a high prognostic value on patient´s overall survival time. Furthermore, TLGwb does not have an inferior prognostic value than MATVwb. CONCLUSION - MATVwb and TLGwb are strong predictors of overall survival in NSCLC patients. Moreover, having the TLGwb more sensibility, by definition, it always provides at least the same prognosis value as MATVwb thus, its use is more accurate, in a clinical perspective.

碩士
國立臺灣大學
醫學工程學研究所
103
Lung cancer is one of the leading cause of cancer deaths worldwide, including Taiwan, across gender. In Taiwan, the male mortality rate of lung cancer is as high as 48.9%. Surgery is the best and effective method in the early stage of the lung cancer. However, only 15% of the diagnosed patients are suitable for early-stage surgery. Surgery only is no longer considered for those being diagnosed at the middle to the late stage. Treatment for patients with middle-stage and late-stage may involve chemotherapy, radiotherapy, or a concurrent chemoradiation therapy. Radiation therapy, which is one of the primary therapeutic approaches for non-small cell lung cancer, is a treatment that uses high-energy rays or particles that destroy lung cancer cells. Radiation-induced lung damage (RILD) is a severe complication of radiotherapy in lung cancer patients that presents as a progressive pulmonary injury affecting prognosis and quality of life in patients. In this study a longitudinal registration algorithm is proposed for evaluating the lung parenchyma change after radiotherapy and the correlation to the given radiation strength and distribution of dosage. The proposed registration algorithm overcomes the large parenchyma change which makes the registration much harder by using anatomical structures around the lung, including using spine for the reference set of rigid registration step; using three anatomical structures: bone structures surface, including sternal, rib and spine, airway wall and surface of lower lung to describe the longitudinal difference of breath holding degree. Reference points are sampled from these three anatomical feature structures for the further step of coherent point set registration. Registered by the proposed longitudinal registration algorithm developed by this study, the correlation of regional dose distribution with longitudinal parenchyma change has been evaluated and obvious parenchyma change in the region of radiation dosage above 22 Gy and in 3~7 month is observed.

Im Rahmen der vorliegenden Dissertationsschrift „Semi-quantitative röntgentomographische Untersuchungen zur Biodistribution von magnetischen Nanopartikeln in biologischem Gewebe“ wurden tomographische Untersuchungen an biologischen Objekten durchgeführt. Bei diesen Objekten handelt es sich um Gewebeproben nach minimal-invasiven Krebstherapien wie zum Beispiel magnetischem Drug Targeting und magnetischer Wärmebehandlung. Der Erfolg dieser Therapien ist sowohl abhängig von der korrekten Verteilung der magnetischen Nanopartikel als auch von der Tatsache, dass diese in der Zielregion in einer ausreichenden Menge vorhanden sind. Das Vorliegen dieser beiden Voraussetzungen ist in der vorliegenden Arbeit untersucht worden. Dabei lag der Schwerpunkt der Arbeit auf der Quantifizierung von magnetischem Material in unterschiedlichen biologischen Gewebeproben mittels Röntgenmikrocomputertomographie (XµCT). Für diesen Zweck wurde ein Kalibrationssystem mit speziellen Phantomen entwickelt, mit dessen Hilfe eine Nanopartikelkonzentration einem Grauwert voxelweise zugewiesen werden kann. Mit Hilfe der Kalibration kann der Nanopartikelgehalt sowohl in monochromatischen als auch in polychromatischen tomographischen Daten im Vergleich zu magnetorelaxometrischen Ergebnissen mit wenigen Prozent Abweichung ermittelt werden. Trotz Polychromasie und damit einhergehenden Artefakten können 3-dimensionale röntgentomographische Datensätze mit einer geringfügigen Konzentrationsabweichung im Vergleich zur quantitativen Messmethode Magnetorelaxometrie semi-quantitativ ausgewertet werden.
The success of the minimal invasive cancer therapies, called magnetic drug targeting and magnetic heating treatment, depends strongly on the correct distribution of the magnetic nanoparticles on one side. On the other side it depends on the fact that a sufficient amount of magnetic nanoparticles carrying drugs is accumulated in the target region. To study whether these two requirements are fulfilled motivates this PhD thesis „Semi-quantitative X-ray-tomography examinations of biodistribution of magnetic nanoparticles in biological tissues“. The analysis of the distribution of the magnetic nanoparticles in tumours and other tissue examples is realized by means of X-ray-micro computer tomography (XμCT). The work focuses on the quantification of the magnetic nanoparticles in different biological tissue samples by means of XµCT. A calibration of the tomographic devices with adequate phantoms, developed in the frame of this work, opens now the possibility to analyze tomographic data in a semi-quantitative manner. Thus, the nanoparticle concentration can be allocated voxel-wise to the grey values of the three-dimensional tomographic data. With the help of calibration of the tomography equipments used, polychromatic as well as monochromatic three-dimensional representations of objects can be analyzed with regard to the biodistribution of magnetic nanoparticles as well as with regard to their quantity. The semi-quantitative results have been compared with results obtained with a quantitative measurement method magnetorelaxometry (MRX). Thereby a good agreement of the semi-quantitative and quantitative data has been figured out.