Abstract The purpose of this study was to evaluate the value of 0.23T low-field magnetic resonance imaging (MRI) and nanocolloid (NC) scintigraphy in assessing joint pathology associated with rheumatoid arthritis (RA). Fat suppression methods combined with contrast media enhancement aid in distinguishing enhancing inflamed tissue from the surrounding fat, especially in the imaging of arthritic joints. The feasibility and image quality of a phase-shift water-fat MRI method for fat suppression at low-field 0.23T open configuration MR scanner was evaluated. The technique was combined with contrast-enhanced imaging to assess the conspicuity of synovial hypertrophy in the joints of 30 RA patients. Improved conspicuity and delineation of synovitis was detected with this method. However, because of a great amount of manual post processing, future development is needed to make this method more feasible. Contrast-enhanced MRI and NC scintigraphy may provide objective and quantitative information about the inflammatory activity in arthritic joints. The value of quantitative and semiquantitative measures of inflammation derived from NC scintigraphy and low-field MRI of the wrist joint of 28 early RA patients was evaluated. Furthermore, it was investigated whether these parameters have predictive value of further erosive development during two years of follow-up. Strong correlations were detected between the NC scintigraphy and MRI measures, and these parameters were associated with laboratory markers of inflammation. During the two-year follow-up, the initial MRI and NC scintigraphy measures were closely related with the progression of wrist joint erosions. Small erosive-like bone defects can occasionally be found in wrist MRI of patients without clinically overt arthritis. The prevalence of these lesions was studied in bilateral wrist MRI examinations of 31 healthy persons. Small lesions resembling erosions were detected in 14 out of 31 subjects. Altogether 24 of the 930 wrist bones evaluated showed such lesions (3%). Thus small changes resembling erosions can be found in the wrist MRI of healthy subjects; the significance of these findings must always be interpreted with reference to the clinical picture. In conclusion, early RA patients with high local inflammatory activity, as detected by NC scintigraphy and MRI are at risk of developing further bone damage. Furthermore, in the follow-up of early RA patients, if clinically sustained response is not achieved, these methods help to identify patients who need more intensive drug treatment.

Os tempos de relaxação longitudinal (T1) e transversal (T2) têm sido frequentemente utilizados na RMN no domínio tempo (RMN-DT) para a obtenção de propriedades físicas e químicas de alimentos, polímeros, petróleo e produtos farmacêuticos. A sequência de pulsos Carr-Purcell- Meiboom-Gill (CPMG) é o método tradicional e mais utilizado para medida de T2, por ser uma técnica rápida para determinação deste tempo de relaxação. Por outro lado, a determinação do T1 não é muito comum, pois as técnicas tradicionalmente utilizadas para sua mensuração, Inversão Recuperação (IR) e Saturação Progressiva (SP), requerem longo tempo de análise. Desta forma, vários autores vêm desenvolvendo métodos bidimensionais (2D) rápidos, para medida de ambos os tempos de relaxação ou da razão T1/T2. Entretanto, estes métodos requerem sequências de pulsos e tratamentos de dados mais sofisticados do que os métodos unidimensionais (1D). Recentemente têm sido desenvolvidas técnicas 1D rápidas para determinação simultânea de T1 e T2 no estado estacionário de precessão livre (Steady State Free Precession - SSFP). Estas sequências são denominadas de Continuous Wave Free Precession (CWFP) e Carr-Purcell Continuos Wave Free Precession (CP-CWFP). A desvantagem desses métodos, é que estes necessitam o ajuste da frequência de offset e do tempo entre os pulsos (Tp), a fim de se obter um ângulo de precessão múltiplo ímpar de π. Desta forma, neste estudo foi proposta a utilização de técnicas provenientes da CWFP com alternância de fase (AF) de π entre os pulsos, as quais na condição de ressonância são independentes do Tp, para Tp<T2∗. Estas sequências são a CWFPx-x e CP-CWFPx-x, que se diferem da CWFP e CP-CWFP somente pela AF de π entre os pulsos de refocalização (x\' e -x\') e a CPMG90y-y que é similar a CPMG90, porém, os pulsos de refocalização são com alternância de fase (y e -y). Nestas sequências, os tempos de relaxação são determinados através da amplitude do sinal logo após o primeiro pulso, |M0|, a amplitude do sinal no estado estacionário, |MEE|, e a constante de tempo exponencial T∗. As constantes de tempo T1 e T2 obtidas com as sequências com AF de π foram iguais ou muito similares às obtidas pelos métodos padrão de determinação destas constantes de tempo (IR e CPMG). Entretanto, dentre as sequências com e sem AF de π apresentadas neste estudo, a CP-CWFPx-x demonstrou ser a melhor sequência para determinação de T1 e T2. Já que o sinal desta apresenta elevada faixa dinâmica do estado quase-estacionário ao estado estacionário que permite a determinação dos tempos de relaxação em amostras com qualquer razão T1/T2. Ainda, esta característica da CP-CWFPx-x possibilita sua aplicação em casos onde há baixa razão sinal ruído. / Longitudinal (T1) and transverse (T2) relation times have been often used in time domain NMR (TD-NMR) to obtain physical and chemical properties of the food, polymers, petroleum and pharmaceutical materials. The Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence is the standard method to measure T2 because it is a fast and accurate method. On the other hand, pulse sequences for T1 determination, as Inversion Recovery (IR) e Saturation Progressive (SP), is not common, such require long time. Under these circumstances, several authors developed two-dimensional (2D) fast methods to determine relation times and/or T1/T2 ratio. However, these methods require pulse sequences that are more sophisticated and difficult data processing than one-dimensional (1D) method. This way, recently, fast 1D techniques, based on Steady State Free Precession (SSFP), has been developed to determine simultaneous T1 and T2.These methods are named the Continuous Wave Free Precession (CWFP) and Carr-Purcell Continuos Wave Free Precession (CP-CWFP). The main disadvantage of these methods is that they require the adjustment of the offset frequency and the time between pulses (Tp) in order to obtain a precession angle odd multiple of π. Therefore, in this study, it proposed to use CWFP technique with π phase alternation (PA), which in resonance are Tp independent, when Tp<T2∗. These pulse sequences were named CWFPx-x, CP-CWFPx-x. They are similar to CWFP and CP-CWFP, however with π PA between refocusing pulse (x\' and -x\'). In addition, other pulse sequence, named CPMG90y-y will be analyzed. It is similar to CPMG90, however, the refocusing pulses have PA (y\' and -y\'). In these pulse sequences, the relaxation times are determine by the signal amplitude immediately after the first pulse |M0| and in the stationary state |Mss|, as wells as the exponential time constant T∗. T1 and T2 determine by the π PA sequences were similar those one obtained by standard methods, IR and CPMG. Nevertheless, among the sequences with or without PA, show in this study, the CP-CWFPx-x proved be the best sequence for T1 and T2 determination, because of its highest dynamic range of the quasi-stationary state to stationary state. Thus, it provides the relaxation times determination independent of the T1/T2 ratio and can be use in cases of the low-reason signal to noise ratio.

Os tempos de relaxação longitudinal (T1) e transversal (T2) têm sido frequentemente utilizados na RMN no domínio tempo (RMN-DT) para a obtenção de propriedades físicas e químicas de alimentos, polímeros, petróleo e produtos farmacêuticos. A sequência de pulsos Carr-Purcell- Meiboom-Gill (CPMG) é o método tradicional e mais utilizado para medida de T2, por ser uma técnica rápida para determinação deste tempo de relaxação. Por outro lado, a determinação do T1 não é muito comum, pois as técnicas tradicionalmente utilizadas para sua mensuração, Inversão Recuperação (IR) e Saturação Progressiva (SP), requerem longo tempo de análise. Desta forma, vários autores vêm desenvolvendo métodos bidimensionais (2D) rápidos, para medida de ambos os tempos de relaxação ou da razão T1/T2. Entretanto, estes métodos requerem sequências de pulsos e tratamentos de dados mais sofisticados do que os métodos unidimensionais (1D). Recentemente têm sido desenvolvidas técnicas 1D rápidas para determinação simultânea de T1 e T2 no estado estacionário de precessão livre (Steady State Free Precession - SSFP). Estas sequências são denominadas de Continuous Wave Free Precession (CWFP) e Carr-Purcell Continuos Wave Free Precession (CP-CWFP). A desvantagem desses métodos, é que estes necessitam o ajuste da frequência de offset e do tempo entre os pulsos (Tp), a fim de se obter um ângulo de precessão múltiplo ímpar de π. Desta forma, neste estudo foi proposta a utilização de técnicas provenientes da CWFP com alternância de fase (AF) de π entre os pulsos, as quais na condição de ressonância são independentes do Tp, para Tp<T2∗. Estas sequências são a CWFPx-x e CP-CWFPx-x, que se diferem da CWFP e CP-CWFP somente pela AF de π entre os pulsos de refocalização (x\' e -x\') e a CPMG90y-y que é similar a CPMG90, porém, os pulsos de refocalização são com alternância de fase (y e -y). Nestas sequências, os tempos de relaxação são determinados através da amplitude do sinal logo após o primeiro pulso, |M0|, a amplitude do sinal no estado estacionário, |MEE|, e a constante de tempo exponencial T∗. As constantes de tempo T1 e T2 obtidas com as sequências com AF de π foram iguais ou muito similares às obtidas pelos métodos padrão de determinação destas constantes de tempo (IR e CPMG). Entretanto, dentre as sequências com e sem AF de π apresentadas neste estudo, a CP-CWFPx-x demonstrou ser a melhor sequência para determinação de T1 e T2. Já que o sinal desta apresenta elevada faixa dinâmica do estado quase-estacionário ao estado estacionário que permite a determinação dos tempos de relaxação em amostras com qualquer razão T1/T2. Ainda, esta característica da CP-CWFPx-x possibilita sua aplicação em casos onde há baixa razão sinal ruído. / Longitudinal (T1) and transverse (T2) relation times have been often used in time domain NMR (TD-NMR) to obtain physical and chemical properties of the food, polymers, petroleum and pharmaceutical materials. The Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence is the standard method to measure T2 because it is a fast and accurate method. On the other hand, pulse sequences for T1 determination, as Inversion Recovery (IR) e Saturation Progressive (SP), is not common, such require long time. Under these circumstances, several authors developed two-dimensional (2D) fast methods to determine relation times and/or T1/T2 ratio. However, these methods require pulse sequences that are more sophisticated and difficult data processing than one-dimensional (1D) method. This way, recently, fast 1D techniques, based on Steady State Free Precession (SSFP), has been developed to determine simultaneous T1 and T2.These methods are named the Continuous Wave Free Precession (CWFP) and Carr-Purcell Continuos Wave Free Precession (CP-CWFP). The main disadvantage of these methods is that they require the adjustment of the offset frequency and the time between pulses (Tp) in order to obtain a precession angle odd multiple of π. Therefore, in this study, it proposed to use CWFP technique with π phase alternation (PA), which in resonance are Tp independent, when Tp<T2∗. These pulse sequences were named CWFPx-x, CP-CWFPx-x. They are similar to CWFP and CP-CWFP, however with π PA between refocusing pulse (x\' and -x\'). In addition, other pulse sequence, named CPMG90y-y will be analyzed. It is similar to CPMG90, however, the refocusing pulses have PA (y\' and -y\'). In these pulse sequences, the relaxation times are determine by the signal amplitude immediately after the first pulse |M0| and in the stationary state |Mss|, as wells as the exponential time constant T∗. T1 and T2 determine by the π PA sequences were similar those one obtained by standard methods, IR and CPMG. Nevertheless, among the sequences with or without PA, show in this study, the CP-CWFPx-x proved be the best sequence for T1 and T2 determination, because of its highest dynamic range of the quasi-stationary state to stationary state. Thus, it provides the relaxation times determination independent of the T1/T2 ratio and can be use in cases of the low-reason signal to noise ratio.

Die Darstellung der Seitenbänder des Hufgelenkes und ihrer Erkrankungen mit Hilfe der Magnetresonanztomographie am stehenden Pferd Mit Einführung des Hallmarq Equine Limb Scanner® wurde es möglich, die distalen Gliedmaßenabschnitte des Pferdes und speziell den Huf einer magnetresonanztomographischen Untersuchung am stehend sedierten Tier zu unterziehen. Die Diagnose der Erkrankung der Kollateralbänder des Hufgelenkes wird damit deutlich einfacher möglich, falls der Niederfeldtomograph hierfür geeignet ist. Zur Bewertung der Eignung des Equine Limb Scanner wurden im Zeitraum von 2005 bis 2009 aus eintausend an der Pferdeklinik Bargteheide mittels des Hallmarq Equine Limb Scanner® 0,27 T-Systems im Stehen untersuchten Pferden diejenigen 69 Patienten ausgewählt, welche ausschließlich eine Erkrankung im Bereich der Kollateralbänder des Hufgelenkes aufwiesen. Die Untersuchung erfolgte damit bei gleichzeitiger physiologischer Zug- und Druckbelastung des zu untersuchenden Gliedmaßenabschnittes. Die Beurteilung der Kollateralbänder und ihrer Erkrankungen wurden anhand eines transversalen und frontalen Fast Spin Echo T2 gewichtet durchgeführt. Um den Ursprungsbereich des Kollateralbandes am Kronbein und den Ansatzbereich am Hufbein zu bewerten wurden zusätzlich Gradientenechos T1 und T2* gewichtet, sowie die STIR Sequenz genutzt. Die Verwendung des Fast Spin Echos T2w mit einer Echozeit von TE=84 ms sowie eine fachgerechte Positionierung des Hufes im Magneten diente zum Ausschluss eines möglichen Magic Angle Effect. Das Kollateralband wurde in drei definierte Zonen (Ursprung, Korpus, Insertion) eingeteilt, um die genaue Lokalisation der Läsion zu beschreiben. Dabei konnte beobachtet werden, dass der Kollateralbandursprung (36%) am häufigsten betroffen ist, gefolgt von Schäden in allen drei Zonen (20%) sowie Ursprung und Korpus (20%), Insertion (16%), und Schäden nur am Korpus (3%) und Korpus und Insertion (3%), Ursprung und Insertion (2%). Bei der klinischen Untersuchung fiel auf, dass bei fast allen Patienten die Lahmheit sowohl auf dem harten als auch auf dem weichen Zirkel jeweils auf der inneren Hand am deutlichsten zu erkennen war. Das Phänomen der schlechten Abgrenzung des Kollateralbandes zu seiner Umgebung (poorly defined borders) wurde im transversalen und frontalen Fast Spin Echo T2w beurteilt. Hierbei konnte festgestellt werden, dass diese Veränderung hauptsächlich im Ursprungsbereich vorkommt. Bei 70% der Patienten mit einem lateralen Kollateralbandschaden im Ursprungsbereich am Kronbein und bei 59% der Patienten mit einem medialen Kollateralbandschaden im Ursprungsbereich am Kronbein konnte dies beobachtet werden. Zur Feststellung einer knöchernen Beteiligung von Kron- und/oder Hufbein wurden zusätzlich Gradientenechos T1 und T2*w sowie die STIR Sequenz hinzugezogen. Dabei zeigte sich, dass eine Beteilung des Knochens vorwiegend im Bereich der medialen Insertion (53%) und weniger lateral (27%) vorkommt. Im Ursprungsbereich ist diese lateral in 16% der Fälle und medial in 3% der Fälle vorzufinden. Die Durchmesser des medialen und lateralen Kollateralbandes im Ursprungs-, Korpus-, und Insertionsbereich wurden im Bild eines transversalen Fast Spin Echo T2w ermittelt. Dazu wurden zusätzlich Referenzwerte von 23 klinisch und magnetresonanztomographisch gesunden Patienten angefertigt. Es wurde festgestellt, dass die Mittelwerte der Flächen der medialen Kollateralbänder in allen drei Zonen bei erkrankten Patienten statistisch signifikant größer waren als bei gesunden Pferden (Ursprung p=0,0297, Korpus p=0,00285 und Insertion p=0,0001603). Die Mittelwerte der Flächen der lateralen Kollateralbänder unterschieden sich bei gesunden und kranken Patienten im Ursprungs-, und Korpusbereich nicht signifikant (Ursprung p=0,4372, Korpus p=0,6258), im Bereich der lateralen Insertion konnte ein deutlicher aber statistisch nicht signifikanter Unterschied (p=0,0753) beobachtet werden bei einer Irrtumswahrscheinlichkeit von α=0,05. Die Ergebnisse dieser Arbeit untermauern die Wichtigkeit moderner bildgebender Verfahren zur Ergänzung und Vervollständigung der klinischen Untersuchung, der Diagnosefindung, der differentialdiagnostischen Abgrenzung, der Therapieplanung und prognostischer Aussagen. Insbesondere im Hufbereich liefert die Magnetresonanztomographie wichtige Informationen zur Erkennung der erkrankten Struktur. Sie bietet eine differenzierte Abgrenzung einer erkrankten Struktur, in diesem Fall der Seitenbänder, die bisher häufig dem Begriff Hufrollensyndrom zugeordnet wurde und mit herkömmlichen bildgebenden Verfahren nicht oder nur teilweise dargestellt werden konnte. Die Untersuchung mit dem Hallmarq Equine Limb Scanner® Niederfeldtomographen am stehend sedierten Pferd ist hierfür nach vorliegenden Ergebnissen hervorragend geeignet.

The aim of this thesis is to perform Magnetic Resonance Imaging at very low field (from 1 mT to 10 mT). A new kind of sensor called "mixed sensor" has been used to achieve a good detectivity at low frequencies. Combining a superconducting loop and a giant magnetoresistance, those detectors have a competitive equivalent field noise compared to existing devices (Tuned coils, SQUIDs and Atomic Magnetometers). They have been combined with flux transformers to increase the coupling between the sample and the sensor. A complete study has been performed to adapt it to mixed sensors and then maximize the gain. This set has been incorporated in an existing small MRI device to test its robustness in real conditions. In parallel, several MRI sequences (GE, SE, FLASH, EPI, ...) have been integrated and adapted to very low field requirements. They have been used to perform in-vivo three dimensional imaging and relaxometry studies on well known products to test their reliability. Finally, a larger setup adapted for full-head imaging has been designed and built to perform images on a larger working volume.

L’enjeu principal de cette thèse a été de démontrer la faisabilité de l’Imagerie par Résonance Magnétique à très bas champ (entre 1 mT et 10 mT). Pour ce faire, un nouveau type de capteur, appelé “capteur mixte”, a été utilisé. Ce détecteur est le résultat de l’association d’une magnétorésistance géante avec une boucle supraconductrice. Il génère un bruit en champ comparable aux détecteurs les plus utilisés dans cette gamme de fréquence (Bobine accordées, SQUIDs, Magnétomètres atomique optique). Le couplage entre l’échantillon observé et le capteur mixte a été grandement amélioré à travers l’utilisation d’un transformateur de flux. Cet intermédiaire a été conçu et optimisé pour maximiser la sensibilité en champ du “capteur mixte”. Cet ensemble a ensuite été introduit dans un IRM à très bas champ magnétique pour tester son efficacité in-situ. Parallèlement, différentes séquences d’IRM (GE, SE, FLASH, EPI,...) ont été développées spécifiquement pour le très bas champ. Elles ont été utilisées pour réaliser de l’imagerie tridimensionnelle in-vivo ainsi que des études relaxométriques sur divers produits. Enfin, un système d’IRM “tête entière” a été construit pour permettre l’acquisition d’images à très bas champ magnétique sur un large volume. / The aim of this thesis is to perform Magnetic Resonance Imaging at very low field (from 1 mT to 10 mT). A new kind of sensor called “mixed sensor” has been used to achieve a good detectivity at low frequencies. Combining a superconducting loop and a giant magnetoresistance, those detectors have a competitive equivalent field noise compared to existing devices (Tuned coils, SQUIDs and Atomic Magnetometers). They have been combined with flux transformers to increase the coupling between the sample and the sensor. A complete study has been performed to adapt it to mixed sensors and then maximize the gain. This set has been incorporated in an existing small MRI device to test its robustness in real conditions. In parallel, several MRI sequences (GE, SE, FLASH, EPI, ...) have been integrated and adapted to very low field requirements. They have been used to perform in-vivo three dimensional imaging and relaxometry studies on well known products to test their reliability. Finally, a larger setup adapted for full-head imaging has been designed and built to perform images on a larger working volume.

付記する学位プログラム名: 京都大学卓越大学院プログラム「先端光・電子デバイス創成学」 / 京都大学 / 新制・課程博士 / 博士(工学) / 甲第23205号 / 工博第4849号 / 新制||工||1757(附属図書館) / 京都大学大学院工学研究科電気工学専攻 / (主査)教授 小林 哲生, 教授 松尾 哲司, 特定教授 中村 武恒 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

Thesis (MScEng (Electrical and Electronic Engineering))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Assessing the quality of fruit has become vitally important for farmers and growers. With retailers placing ever stricter requirements on fresh produce, growers have to spend a greater amount of time and effort sorting and grading their harvest. Increasingly, vendors are placing requirements not only on external factors like colour, and firmness, but on internal quality factors such as sugar content and acidity, because, although consumers buy fruit based on their external appearances, the taste of the fruit is what determines whether the consumer will buy again. Different techniques exist that probe the internal quality of fruit non-destructively. The technique most widely used today is Near Infrared Spectroscopy. This technique is powerful, but has certain limitations such as poor reliability and the need for constant recalibration. This thesis suggests an alternative method for evaluating internal fruit quality based on low-field nuclear magnetic resonance detected by superconducting quantum interference devices. It introduces the theory of SQUIDs and NMR, and evaluates the use of SQUID detected NMR spectroscopy as a method for determining the internal quality of fruit. The fabrication techniques and processes are explained in detail and a design for a SQUID detected NMR spectrometer is given. Relevant simulations and simulation results are also given. No working SQUID could be fabricated and, as such, no working NMR spectrometer was demonstrated. This thesis serves as a reference work for future research. / AFRIKAANSE OPSOMMING: Bepaling van die gehalte van vrugte het vir boere uiters belangrik geword. Met kleinhandelaars wat al strenger vereistes plaas op vars produkte moet boere meer tyd en inspanning bestee met die sortering en gradering van hul oes. Handelaars plaas nie net vereistes op eksterne kwaliteitsfaktore soos kleur en fermheid nie, maar begin al hoe strenger vereistes plaas op interne kwaliteitsfaktore soos suikerinhoud en suurgehalte, want, hoewel verbruikers vrugte koop op grond van hul eksterne kwaliteitsfaktore, is dit die smaak van die vrug wat bepaal of die verbruiker weer die vrug sal koop. Verskillende tegnieke bestaan wat die interne kwaliteit van vrugte op ’n nie-destruktiewe manier kan bepaal. Die mees algemene tegniek is Naby Infrarooi Spektroskopie. Hierdie tegniek is kragtig maar het sekere beperkings soos swak betroubaarheid en die noodsaaklikheid van konstante herkalibrasie. Hierdie tesis stel ’n alternatiewe metode vir die evaluering van interne vrugkwaliteit gebaseer op lae-veld kernmagnetiese resonans waargeneem deur supergeleidende kwantum inmenging toestelle voor. Dit stel die teorie van SKWITs en KMR bekend, en evalueer die gebruik van SKWIT-bespeurde KMR spektroskopie as ’n metode vir die bepaling van die interne kwaliteit van vrugte. Die fabrikasie tegnieke en prosesse word in detail verduidelik en ’n ontwerp vir ’n SWKIT opgevangde KMR spektrometer word gegee. Toepaslike simulasies en simulasie resultate word ook gegee. Geen werkende SKWIT kon vervaardig word nie en as gevolg daarvan kon geen werkende KMR spektrometer gedemonstreer word nie. Hierdie tesis dien as ’n naslaan werk vir toekomstige navorsing.

A ressonância magnética nuclear no domínio do tempo (RMN-DT), possui inúmeras vantagens em relação à RMN de alta resolução tais como: baixo custo do equipamento, de análise e manutenção e não precisa de operador especializado. Apesar disso a técnica de RMN-DT ainda tem um campo muito grande de desenvolvimento para análise de componente com baixa mobilidade molecular como sólidos amorfos e cristalinos. Isto requer o desenvolvimento de instrumentação e/ou métodos de análise que se consiga medir o sinal de RMN em menos de 10 µs, após a irradiação da amostra. O tempo entre o final da excitação e o começo da aquisição do sinal é conhecido como tempo morto. A opção estudada nesta dissertação para redução do tempo morto foi o uso de pulsos formatados, que já são utilizados em experimentos de RMN em alta resolução e de imagens por RMN, para excitação seletiva. No entanto, pulsos formatados não têm sido utilizados em RMN-DT. A principal vantagem desses pulsos é que eles são projetados para que a potência no final do pulso seja mínima, o que diminui o tempo morto do espectrômetro. No caso do pulso convencional, na forma retangular, a potência no final do pulso é máxima, o que aumenta o tempo morto. Os pulsos formatados utilizados foram pulsos Sinc, Sine e também pulsos com início similar aos pulsos retangulares, com potência máxima, e que depois de certo tempo tem decaimento similar aos pulsos Sinc e Sine. Esses pulsos inéditos foram denominados Reta-Sine e Reta-Sinc. Foram avaliados amostras heterogêneas (solido-líquido) e sólidas como CaSO4.2H2O, farelo de soja, polímeros amido e Policaprolactona. Esses pulsos também foram usados para monitorar reações de polimerização da resina epóxi com poliamida (Araldite®). Com esses métodos foi possível reduzir o tempo morto de espectrômetro de 23MHz 23 µs para 14 µs e de 20 Mhz de 9 para 6 µs. Com isso conclui-se que o pulsos formatados podem reduzir significativamente o tempo morto de espectrômetro de RMN-DT, sem acessórios ou custo financeiro, pois as sequências podem ser instalada e usadas em qualquer espectrômetro moderno. / The nuclear magnetic resonance in the time domain (TD-NMR) It has numerous advantages over the high-resolution NMR such as low cost of equipment, analysis and maintenance and requires no specialized operator. Nevertheless NMR DT technique also has a very large development field for component analysis with a low molecular mobility as amorphous and crystalline solids. This requires the development of instrumentation and / or analysis methods that can measure the NMR signal in less than 10 µs, after irradiation of the sample. The time between the end of excitation and signal acquisition the beginning is known as dead time. The option studied in this work to reduce the dead time was the use of formatted pulses, which are already used in NMR experiments in high resolution and NMR imaging for selective excitation. However, formatted pulses have not been used in NMR DT. The main advantage of these pulses is that they are designed so that the power in the pulse end is minimized, which reduces the dead time of the spectrometer. In the case of conventional pulse in rectangular form, the power in the pulse end is maximum, which increases the dead time. Formatted pulses used were Sync pulses, Sine and also pulses beginning similar to rectangular pulses with maximum power, and after a while has similar decay to Sync and Sine wrists. These unpublished wrists were called Sine and Straight-Line-Sync. They were evaluated heterogeneous samples (solid-liquid) and solid as CaSO4.2H2O, soybean meal, starch polymers and Polycaprolactone. These pulses are also used to monitor polymerization reactions the polyamide epoxy resin (Araldite®). With these sequences it was possible to reduce the dead time of 23 µs 23MHz spectrometer to 14 µs e 20 Mhz from 9 to 6 µs. With this it is concluded that the formatted pulses can significantly reduce the dead time DT-NMR spectrometer, without fittings or financial cost because the sequences can be installed and used in any modern spectrometer.

L'objectif de ces travaux a été de démontrer l'intérêt d'appliquer la Résonance Magnétique Nucléaire (RMN) à des problématiques analytiques dans le domaine des produits de santé. Deux axes principaux ont été abordés. Le premier cible principalement les compléments alimentaires adultérés. Les études ont été menées par RMN à haut champ. Diverses expériences de RMN mono et bidimensionnelles ont permis de mettre en évidence la présence illicite, dans des compléments alimentaires déclarés comme naturels, de médicaments à visée érectile et/ou de leurs analogues structuraux. Ces substances actives peuvent induire, à l'insu du consommateur, des risques importants pour sa santé. De plus, des analyses menées par RMN 1H et 13C ont permis de révéler des cas d'adultération de compléments alimentaires pour sportifs par des substances considérées comme dopantes par la règlementation en vigueur. Le second axe a été mené en explorant les potentialités de la RMN à bas champ pour l'analyse de médicaments. Cette technique, moins onéreuse et plus adaptée à des analyses de routine que la RMN à haut champ, s'est avérée très efficace pour mettre en évidence des cas de médicaments falsifiés par de simples expériences RMN 1H monodimensionnelles. Des expériences bidimensionnelles de diffusion (DOSY : Diffusion Ordered SpectroscopY), qui permettent d'obtenir la séparation virtuelle des constituants d'un mélange complexe, ont été mises au point et appliquées, pour la première fois, à la dé-formulation d'un médicament. / The purpose of this research was to demonstrate the interest of Nuclear Magnetic Resonance (NMR) to deal with analytical issues in the field of health products. The studies focused on two main axes. The first one targeted adulterated dietary supplements and the studies were conducted with high-field NMR. Various mono and bidimensional experiments revealed the illegal presence in dietary supplements claimed as natural, of erectile drugs and/or their structural analogues. These active substances can induce, without the consumer's knowledge, significant risks to his health. Moreover, analyses conducted with 1H and 13C NMR, highlighted cases of adulteration of dietary supplements for exercise and athletic performance with doping substances. The second axis explored the potential of low-field NMR for drug analysis. This technique, which is less expensive and more suitable for routine analyses than high-field NMR, has proved very effective in detecting cases of falsified medicines by simple one-dimensional 1H NMR experiments. Two-dimensional diffusion experiments (DOSY: Diffusion Ordered Spectroscopy), which provide virtual separation of the constituents of a complex mixture, were developed and applied for the first time to the de-formulation of a drug.