L'imagerie par résonance magnétique cinétique en contraste de phase (IRM-PC) sensible aux flux permet la quantification et la caractérisation des écoulements sanguins in-vivo de manière non-invasive. En clinique, cette séquence est principalement appliquée en une coupe à deux dimensions (2D), mais elle a évolué vers une acquisition volumique des vitesses (4D IRM-PC) offrant une quantification complète de l'écoulement dans l'ensemble d'une arborescence vasculaire. Cependant, cette technique nécessite d'outils de post-traitement qui sont encore peu nombreux et d'une évaluation de la précision de la mesure des vitesses en 2D et 4D dans des vaisseaux de petits diamètres. Dans ce travail de thèse, un logiciel de post-traitement des images 2D et 4D IRM-PC a été mis au point avec une interface facile et est utilisé par des équipes de recherches et des cliniciens. Des protocoles d'acquisitions des vitesses par IRM pour des vaisseaux constitués de diamètres millimétriques ont été développés. Ces protocoles ont été validés in-vitro à partir d'un travail réalisé sur fantôme qui montre une erreur inférieure à 10% dans la précision des mesures. Les protocoles validés ont été appliqués sur les artères de la région cervico-faciale chez 30 volontaires sains afin de constituer la première base hémodynamique des vaisseaux de cette région. Des patients atteints de pathologies cervico-faciale ont bénéficié de ce protocole et son intérêt a été montré dans le suivi du patient et pour la planification chirurgicale / Phase-Contrast Magnetic Resonance imaging (PC-MRI) is a non-invasive technique used for quantification and characterization of the blood flow. In clinical pratice, this sequence is principally used in a two-dimensional single slice (2D), but it evolved to a velocity volumic acquisition (4D PC-MRI) allowing a complete quantification of the flow through a vascular tree. However, this technique requires post-processing software which are few and an evaluation of the velocity accuracy in 2D and 4D to quantify vessels with millimetric diameter. In this thesis work, a user-friendly post-processing software of 2D and 4D MRI images has been developed and it is used by research teams and clinicians. MRI acquisitions protocols of velocities for vessels composed by millimetric has been created. These protocols developed in-vitro form a phantom work show an error in the measurement accuracy less than 10%. These validated protocols are applied in facial area arteries in 30 healthy volunteers to create the first hemodynamic data base of arteries in this area. Then, the protocols have been performed to patients with pathologies affecting the facial area to show is interest on patient follow-up and surgical treatment

Die Kombination von Positronen-Emissions-Tomographie (PET) und Röntgen-Computertomographie (CT) in Form moderner PET/CT-Geräte ermöglicht die Nutzung der CT-Information zur Korrektur der Photonenschwächung in der PET. Allerdings können Bewegungen, die zum Beispiel durch die Atmung hervorgerufen werden können, zu einer fehlerhaften Schwächungskorrektur führen. Die Einführung von zeitlich aufgelöster Bildgebung für beide Modalitäten (4D-PET/4D-CT) ermöglicht nicht nur die Auflösung von periodischen Bewegungen, sondern auch die Reduktion dieser Fehler in der Schwächungskorrektur. Dazu werden die einzelnen Datensätze des 4D-PET, die jeweils einer bestimmten Bewegungsphase entsprechen, mit dem entsprechenden CT-Datensatz dieser Atemphase schwächungskorrigiert. In der vorliegenden Arbeit wurde diese phasenkorrelierte Schwächungskorrektur des 4D-PET mit dem 4D-CT am Universitästsklinikum Dresden installierten PET/CT ermöglicht und anhand von Phantomexperimenten mit anderen Schwächungskorrekturmethoden für 4D-PET verglichen. Dazu musste zunächst die Aufnahme von 4D-CT an dem verwendeten PET/CT ermöglicht und dessen Synchronität mit dem 4D-PET hergestellt werden. Außerdem wurde ein vorhandenes Atemphantom so modifiziert, dass es typische Bewegungen von Bronchialkarzinomen in zwei Dimensionen und mit zwei möglichen Atemmustern simuliert. Die phasenkorrelierte Schwächungskorrektur führte zu einer quantitativ korrekten Wiederherstellung des Aktivitätsvolumens, der darin enthaltenen Aktivität sowie der Bewegungsamplitude und stellt somit die beste der hier verglichenen 4D-PET-Schwächungskorrekturmethoden dar. Diese Ergebnisse lassen vermuten, dass die phasenkorrelierte Schwächungskorrektur auch bei klinischer Anwendung eine signifikante Verbesserung in oben genannten Punkten darstellt. Dies sollte in Zukunft an Patientendaten überprüft werden. / The combination of Positron Emission Tomography (PET) and Computed Tomography (CT) in one device allows the use of CT-information for attenuation correction in PET. Though motion, for example induced by respiration, can cause inaccurate attenuation correction. The implementation of time-resolved imaging methods for both modalities (4D-PET/4D-CT) enables not only the resolution of motion but also the reduction of artifacts caused by attenuation correction. Therefore, the single datasets of the 4D-PET that are related to a individual respiratory phase, are attenuation corrected with the corresponding dataset of the 4D-CT. This phase correlated attenuation correction of the 4D-PET with the 4D-CT was implemented at the PET/CT installed at the Universitätsklinikum Dresden. For that purpose the acquisition of 4D-CT was implemented at the PET/CT and its synchronisation with the 4D-PET was verified. Furthermore the new attenuation correction method was compared with other attenuation correction methods by performing phantom experiments. Therefore an exisisting respiratory phantom had to be modified to perform typical lung tumor motion in two dimensions with two possible patterns of respiration. The phase correlated attenuation correction leads to a quantitatively correct restauration of the activity volume, its total activity and its motion amplitude. Compared with other correction methods, the phase correlated attenuation correction shows the best results in all examined criteria. This findings suggest that the clinical application of the phase correlated attenuation correction will also lead to a significant improvement in all mentioned points. This has to be verified by analyzing patient data.

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第20222号 / 医博第4181号 / 新制||医||1019(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 鈴木 実, 教授 高田 穣, 教授 武藤 学 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

The accurate assessment of hemodynamics and its parameters play an important role when diagnosing cardiovascular diseases. In this context, 4D Flow Magnetic Resonance Imaging (4D Flow MRI) is a non-invasive measurement technique that facilitates hemodynamic parameter assessment as well as quantitative and qualitative analysis of three-directional flow over time. However, the assessment is limited by noise, low spatio-temporal resolution and long acquisition times. Consequently, in regions characterized by transient, rapid flow dynamics, such as the aorta and heart, capturing these rapid transient flows remains particularly challenging. Recent research has shown the feasibility of machine learning models to effectively denoise and increase the spatio-temporal resolution of 4D Flow MRI. However, temporal super-resolution networks, which can generalize on unseen domains and are independent on boundary segmentations, remain unexplored.  This study aims to investigate the feasibility of a neural network for temporal super-resolution and denoising of 4D Flow MRI data. To achieve this, we propose a residual convolutional neural network (based on the 4DFlowNet from Ferdian et al.) providing an end-to-end mapping from temporal low resolution space to high resolution space. The network is trained on patient-specific cardiac models created with computational-fluid dynamic (CFD) simulations covering a full cardiac cycle. For clinical contextualization, performance is assessed on clinical patient data. The study shows the potential of the 4DFlowNet for temporal-super resolution with an average relative error of 16.6 % on an unseen cardiac domain, outperforming deterministic methods such as linear and cubic interpolation. We find that the network effectively reduces noise and recovers high-transient flow by a factor of 2 on both in-silico and in-vivo cardiac datasets. The prediction results in a temporal resolution of 20 ms, going beyond the general clinical routine of 30-40 ms. This study exemplifies the performance of a residual CNN for temporal super-resolution of 4D flow MRI data, providing an option to extend evaluations to aortic geometries and to further develop different upsampling factors and temporal resolutions. / En noggrann bedömning av hemodynamiken och dess parametrar spelar en viktig roll vid diagnos av kardiovaskulära sjukdomar. I detta sammanhang är 4D Flow Magnetic Resonance Imaging (4D Flow MRI) en icke-invasiv mätteknik som underlättar bedömning av hemodynamiska parametrar samt kvantitativ och kvalitativ analys av flöde. Bedömningen begränsas av brus, låg spatio-temporal upplösning och långa insamlingstider. I områden som karakteriseras av snabb flödesdynamik, såsom aorta och hjärta, är det därför fortfarande särskilt svårt att fånga dessa snabba transienta flöden. Ny forskning har visat att det är möjligt att använda maskininlärningsmodeller för att effektivt reducera brus och öka den spatio-temporala upplösningen i 4D Flow MRI. Nätverk för temporal superupplösning, som kan generaliseras till osedda domäner och är oberoende av segmentering, är fortfarande outforskade.  Denna studie syftar till att undersöka genomförbarheten av ett neuralt nätverk för temporal superupplösning och brusreducering av 4D Flow MRI-data. För att uppnå detta föreslår vi ett residual faltningsneuralt nätverk (baserat på 4DFlowNet från Ferdian et al.) som tillhandahåller en end-to-end-mappning från temporalt lågupplöst utrymme till högupplöst utrymme. Nätverket tränas på patientspecifika hjärtmodeller som skapats med CFD-simuleringar som spänner över en hel hjärtcykel. För klinisk kontextualisering utvärderas nätverkets prestanda på kliniska patientdata. Studien visar potentialen av 4DFlowNet för temporal superupplösning med ett genomsnittligt relativt fel på 16,6 % på en osedd hjärtdomän, vilket överträffar deterministiska metoder som linjär och kubisk interpolation. Vi konstaterar att nätverket effektivt minskar brus och återställer högtransient flöde med en faktor på 2 på både in-silico ochin-vivo hjärtdataset. Förutsägelsen resulterar i en temporal upplösning på 20 ms, vilket är mer än den allmänna kliniska rutinen på 30-40 ms. Denna studie exemplifierar prestandan hos en residual CNN för temporal superupplösning av 4D-flödes-MRI-data, vilket ger möjlighet att utvidga utvärderingarna till aortageometrier och att vidareutveckla olika uppsamplingsfaktorer och temporala upplösningar.

The construction industry leads the statistics when it comes to reports of accidents and serious incidents compared to other industries. Falls from heights, improper working positions as well as heavy lifting and carrying, are some of the most common safety risks. The purpose of this study is to examine the potential of using a 4D-site layout plan at construction sites and analyze its impact on safety planning. Furthermore, the advantages and disadvantages of the developed 4D-site layout plan are compered the conventional 2D-site layout plan. The aim of this thesis is to develop a method that enhances the visualization of the site layout plan and thus increases safety on the construction site. The methods used in this study include semi-structured interviews, field observation to closely study the current project, and ultimately constructing a 4D model of the site layout plan. The interview results indicate that current site layout plans have limited visualization as they depict the construction site in 2D and from a single angle. Respondents emphasize that current site layout plans are used as a basis for discussion for logistics planning but not for safety planning. This suggests that with 2D-site layout plans, it’s challenging to foresee and evaluate potential safety risks that may occur on the construction site. Respondents agreed that the 4D-site layout plan enhances the understanding of the construction site and improves communication among stakeholders, as it clearly illustrates ongoing progress. However, respondents had varying opinions on the usability and implementation of the 4D-site layout plan at the construction site, as it’s difficult to predict the effects this method has on safety planning. According to the literature review, there are numerous advantages to 4Dvisualization, such as the ability to plan and evaluate work processes and more straightforward way to convey results to relevant parties. Through this method, risky situations can be identified and addressed throughout the entire construction process, contributing to a safer construction site. In this thesis, Unity software has been utilized to visualize the site layout plan in 4D, and this method can enhance safety efforts at the construction site. In conclusion, 4D-visualization requires advanced technology and developed expertise within this field. Further investigations and observations are necessary to evaluate this method at construction sites and explore its potential in safety efforts. The work concludes with suggestions for further studies that haven’t been covered in this thesis. One of the presented suggestions is to upgrade the method to 5D by integrating the cost aspect. / Byggbranschen toppar statistiken när det gäller anmälningar av olyckor och allvarliga tillbud jämfört med andra branscher. Fall från högre platser, olämpliga arbetsställningar samt tunga lyft och bärande är några av de vanligaste säkerhetsriskerna. Syftet med detta examensarbete är att utforska potentialen för att använda en 4DAPD-plan, det vill säga en arbetsplatsdispositionsplan, på byggarbetsplatser och analysera dess påverkan på säkerhetsplaneringen. Dessutom undersöks fördelar och nackdelar med den utvecklade APD-planen i jämförelse med den konventionella 2DAPD-planen. Målet med detta examensarbete är att utveckla en metod som förbättrar visualiseringen av APD-planen och därigenom ökar säkerheten på byggarbetsplatsen. De använda metoderna i detta examensarbete inkluderar semi-strukturerade intervjuer, fältobservation för att studera det aktuella projektet närmare och slutligen ett modellbyggande av APD-planen i 4D. Resultat från intervjuerna pekar på att dagens APD-planer har en begränsad visualisering eftersom de visar byggarbetsplatsen i 2D och från en enda vinkel. Respondenterna betonar att dagens APD-plan används som underlag för logistikplanering, men inte för säkerhetsplanering. Detta inkluderar att med 2DAPD-planer är det svårt att förutse och utvärdera potentiella säkerhetsrisker som kan uppstå på byggarbetsplatsen. Respondenterna var eniga om att 4D-APD-planen ökar förståelse för byggarbetsplatsen och förbättrar kommunikationen bland de involverade, eftersom den tydligt visar pågående arbete. Däremot hade respondenterna olika åsikter om användbarheten och tillämningen av 4D-APD-planen på byggarbetsplatsen, då det är svårt att förutse om vilka effekter denna metod har på säkerhetsplaneringen. Enligt litteraturstudien finns det många fördelar med 4D-visualisering, till exempel möjligheten att planera och utvärdera arbetsmoment samt ett enklare sätt att kommunicera resultat till berörda parter. Genom denna metod kan riskfyllda situationer identifieras och åtgärdas under hela byggprocessen, vilket kan bidra till en säkrare byggarbetsplats. I detta examensarbete har Unity programmet använts för att visualisera APD-planen i 4D, och genom denna metod kan säkerhetsarbetet förbättras på byggarbetsplatsen. Slutligen kräver 4D-visualisering avancerad teknik samt djupgående kunskaper och erfarenheter inom området. Det behövs ytterligare utredningar och observationer för att kunna utvärdera denna metod på byggarbetsplatser och utforska dess möjligheter för säkerhetsarbete. Arbetet avslutades med några förslag på fortsatta studier som inte täckts av detta examensarbete. Ett av de presenterade förslagen är att uppgradera metoden till 5D genom att integrera kostnadsaspekten.

In space experiments that study Charged Cosmic Rays (CCRs), a calorimeter combined with a tracker is used to identify the incoming particles. The back-scattered particles from the calorimeter can enter the tracker creating additional hits. The timing measurement for each hit can be utilized to distinguish between the back-scattered and primary hits. Low Gain Avalanche Detector (LGAD) is a promising technology for detecting particles with timing precision O(10) ps being developed for High-Energy Physics (HEP) experiments. With the current LGAD technology, it is difficult to achieve fine segmentation of sensors into channels. The typical size of an LGAD sensor is O(1 mm2) for HEP experiments whereas, silicon strip sensors in space applications have 50-60 cm long channels with 100 μm pitch, resulting in a channel area of about 1 cm2. The goal of this thesis is to evaluate the use of LGADs for timing in space. In view of this, the channel size of LGADs needs to be modified to O(1 cm2) to be utilized in space experiments. The larger channel size poses challenges for the time resolution of the sensors. First, the issues due to signal shape non-uniformity in pads and strips have been investigated, which can affect the timing performance of the sensors. Second, the fill factor problem due to the segmentation of LGADs is studied. The segmentation of LGADs results in the reduced active area of the sensor. Two different LGAD technologies namely RSD and TI-LGADs have been characterized as a possible solution to the segmentation problem. Finally, the work related to the scaling of the channel size of the LGADs from 1 mm2 to 1 cm2 is presented. TCAD simulations have been performed to obtain the process parameters for the gain layer to achieve a gain of 100 thought to improve the time resolution for larger channel areas. A batch of 16 wafers has been produced to study the adaptation of LGADs for space applications. The fabrication parameters have been optimized using TCAD simulations. This thesis contains the first characterization of the batch. This work aims to provide the instruments to advance the understanding of the fundamental properties of the universe and contribute to the development of LGADs technology for astroparticle physics experiments in space.

No description available.

The building sector is in need of enhanced efficiency. The methodology behind Building Information Modeling, BIM, which is being used increasingly in extent, has shown to improve the efficiency of the flow of information in projects. The concept of 4D is based on the working method and methodology of integration between BIM and a time schedule. This can under certain circumstances further improve the efficiency of the planning phase of building- and civil engineering projects. Today there are not many companies that have begun to use this method and methodology, even though it has been shown to be effective for some conditions. This report addresses this dilemma, why and how companies can implement the work and process of 4D.

A leishmaniose é causada por tripanossomatídeos do gênero Leishmania que infectam preferencialmente macrófagos. Vários factores influenciam a forma e a severidade da doença: a espécies de Leishmania e a resposta imune do hospedeiro. Considerando a importância da ativação dos macrófagos na infecção, o potencial papel de CD100 na modulação da ativação dos macrófagos e os nossos dados anteriores de que CD100 solúvel (sCD100) aumenta a infectividade pelo parasita, pretendemos caracterizar os efeitos do CD100 na infecção por Leishmania (L.) amazonensis. Descobrimos que ambos, promastigotas e amastigotas, são mais infecciosos na presença de sCD100 e que o receptor CD72 é o responsável pelo aumento da infecção. Experimentos in vitro indicaram índice de infecção similares entre macrófagos nocautes para CD100 e selvagens, mas curiosamente, os animais nocautes infectados desenvolveram lesões significativamente menores do que os selvagens, sugerindo que sCD100 presente em outras células pode influenciar a formação da lesão. / Leishmaniasis is caused by trypanosomes of the genus Leishmania that preferentially infect macrophages. Several factors influence the form and severity of the disease: the species of Leishmania and the host immune response. Considering the importance of the activation of macrophages in infection, potential role of CD100 in the modulation of macrophage activation and our previous data that CD100 soluble (sCD100) increase the infectivity of the parasite, we intend to characterize the effect of CD100 in infection with Leishmania (L.) amazonensis. We found that both promastigotes and amastigotes, are most infectious in the presence of sCD100 and the CD72 receptor is responsible for the increased infection. In vitro experiments indicated similar infection rate of macrophages to CD100 knockouts and wild type, but interestingly, the infected knockout animals developed significantly smaller lesions than wild type suggesting that sCD100 present in other cells may influence lesion formation.

AB S TRACT purpose : To further investigate the effects of respiratory motions on CBCT imaging, as well as 4D-CT examinations, with a future goal of using obtained results to implement new methods for individual margins and daily matching procedures into routine clinical practice. background : Since the implementation of CBCT combined with modern accelerators, a higher degree of accuracy has been made possible in RT. However, due to the slow gantry speed of linear accelerators, the imaging procedure of CBCT is a slow process which is thereby degraded by internal motion such as respiration. material and methods : Attain patient specific respiratory motion patterns from CBCT projection data of previous examinations. Utilize this data to perform simulations for both CBCT and 4D-CT using a steering system which allows for arbitrary motion patterns in the longitudinal direction. results : Various imaging with CBCT showed that the resulting images during respiratory motion, can be described by the Probability Density Function of the motion for as long as it does not cause related distortions. This also meant that convolution could be implemented as a model to estimate the CBCT images during oscillation, knowing the object and motion pattern. The 4D-CT examinations using the steering system showed that irregular motion patterns were less accurately described than regular patterns, making the actual motion an important feature to combine together with the measured amplitude. conclusions : It was made clear that CBCT images can be described by the PDF, and thus can be seen as a Color Intensity Projection of the object position. Also it has been shown that the projection data of CBCT images contains valuable information about the respiratory motion of the patient. Another conclusion is that with the help of fiducials, the position of the target within the respiratory cycle can be determined relative to the 4D-CT examination, enabling further input data as to the daily matching procedure, proper applied margins as well as dose to the OAR.