Sprachverstehen erfordert ein Zusammenspiel verschiedener mentaler Prozesse. Zuerst muss der kontinuierliche akustische Sprachstrom in einzelne Einheiten (z.B. Wörter) unterteilt werden. Segmentale und suprasegmentale linguistische Information unterstützt den Segmentierungsprozess. Wissen über die neuronalen Mechanismen dieser Prozesse ist wesentlich um Sprachverarbeitungs- und Sprachentwicklungsprozesse zu verstehen. Ziel der Dissertation ist die Charakterisierung neuronale Korrelate der Verarbeitung sprachrelevanter akustischer Information bei Erwachsenen und Säuglingen mittels kombinierter Nahinfrarot-Spektroskopie- und Elektroenzephalographie-Messungen. Studie I untersucht zeitliche und topographische Aspekte der phonotaktischen Verarbeitung bei Erwachsenen. Die Ergebnisse zeigen, dass phonotaktische Information links-hemisphärisch verarbeitet wird. Ob diese Lateralisierung auf die linguistischen oder akustischen Eigenschaften der Reize zurückzuführen ist bleibt unklar. Erwachsenen-Studien zeigen, dass die lateralisierte Verarbeitung von Sprache aus der Spezialisierung des auditorischen Kortex für bestimmte zeitliche akustische Variationen im Sprachsignal resultiert. Studie II und III untersuchen, ob diese Asymmetrie bereits bei Säuglingen vorliegt. Dafür werden zeitlich variierte nicht-sprachliche akustische Reize präsentiert. Die zeitlichen Variationen entsprechen kritischen Modulationen im Sprachsignal. Studie II zeigt, dass bei Neugeborenen die zeitlichen Modulationen zu differenzierten und lateralisierten Verarbeitungsmustern führen, die über die ersten Lebensmonate konstant bleiben (Studie III). Die Ergebnisse unterstützen die Annahme, dass die lateralisierte Verarbeitung von Sprache mit einer Spezialisierung des auditorischen Kortex für bestimmte zeitliche Frequenzen zusammenhängt. Das Gehirn ist von Geburt auf die Wahrnehmung zeitlicher akustischen Variationen spezialisiert, die für die Entschlüsselung des Sprachsignals relevant sind. / The comprehension of spoken language requires the segmentation of the continuous acoustic speech stream into smaller units (e.g., words). Segmental and suprasegmental linguistic information guide the segmentation process. Investigating the underlying neuronal mechanisms is crucial for understanding the general nature of language perception, and language acquisition in infancy. This dissertation aimed to determine neuronal mechanisms underlying the perception of basic auditory cues relevant for the segmentation of speech in adults and infants using concurrent recordings of near-infrared spectroscopy and electroencephalography. Study I assessed temporal and topographic characteristics of phonotactic processing in adults, thus forming the basis for future studies in infants. The results show that phonotactic processing recruits a left hemispheric network. Whether these asymmetries are a function of linguistic attributes or of basic temporal signal properties is under debate. Studies in adults link hemispheric specialization for speech perception to an asymmetry in cortical tuning and reveal that the auditory cortices are differentially sensitive to temporal features of speech. Whether this asymmetry is already established in infancy is addressed by study II and III. These studies used acoustic non-linguistic sounds that vary in their temporal structure, thus sharing critical temporal features with language. Study II reveals that newborns process temporally varying stimuli in a differential and lateralized fashion. Study III indicates that this lateralization pattern remains constant over the first months of life. The findings support the notion that the lateralization of language functions might result from a specialization for different acoustic properties. The data provide further evidence that language acquisition is linked to basic capacities in auditory processing, and reveal that from birth the brain is tuned to critical temporal properties of linguistic signals.

Die Arbeit stellt die optische Methode der Nahinfraotspektroskopie (NIRS) in der Anwendung zur Messung von Oxygenierungsänderungen am Kopf des Erwachsenen dar. Dabei werden zunächst methodische Aspekte behandelt. Dies schließt (i) die Frage ein ob Änderungen des Redox-Status der Cytochromoxidase mit der NIRS bestimmt werden können; (ii) diskutiert die Ergebnisse zur Darstellbarkeit schneller optischer Signale, wie sie in invasiven Studien als Korrelat der elektrophysiologisch messbaren neuronalen Prozesse bekannt sind und (iii) richtet sich auf die bessere Definition des Messvolumens, sowohl im Sinne einer zweidimensionalen Bildgebung der Oxygenierungs-änderungen an der Hirnoberfläche als auch die Möglichkeiten zu einer Tiefenauflösung. Bezüglich der grundlegenden physiologischen Fragestellungen thematisiert die Arbeit im zweiten Teil Fragestellungen der neuro-vaskulären Kopplung. Dies ist die Frage, wie sich die neuronale, elektrophysiologisch meßbare Aktivität von Neuronen in eine hämodynamische Antwort übersetzt. Hier sind die Einordnung der NIRS-Befunde in ein Modell der vaskulären Antwort aber auch gleichzeitige Messungen mit elektrophysiologischen Methoden und weiteren vaskulär-basierten Methoden, insbesondere der funktionellen Kernspintomographie, Fokus. Schließlich stellt die Arbeit die klinischen Perspektiven in der Neurologie dar, wie sie sich besonders für die vaskulären Erkrankungen des Gehirns aber auch für Fragestellungen bei Epilepsie und Migräne ergeben. / We here present a optical method, near-infrared spectroscopy (NIRS), and its application for non-invasive measurement of oxygenation changes in the adult head. The first part deals with methodological aspects of NIRS. This includes (i) the question whether changes of cytochrome-oxidase redox state can be monitored by NIRS; (ii) the discussion of study results investigating the non-invasive detectability of fast optical signals , known to reflect electrophysiological changes in invasive animal preparations; (iii) and finally the issue how the sampling volume can be better defined, and with respect to a two-dimensional imaging approach of cortical oxygenation changes and the methodological options to achieve a depth resolution. In a second chapter we then focus on the underlying physiological issues mainly raising questions concerning neuro-vascular coupling. Neuro-vascular coupling means the translation of neuronal activity, as can be measured by electrophysiological methods, into the haemodynamic response. Here the focus is on how the results of functional NIRS studies can be interpreted in the light of a model of the vascular response. Also the simultaneous assessment of electrophysiological parameters or the parameters of other methods based on the vascular response, especially fMRI, are dealt with in this chapter. Finally the here presented work gives perspectives of potential clinical applications of the method. This primarily means the application in stroke and other vascularly diseases in neurology, but also includes questions in the clinical fields of epilepsy and migraine.

In this thesis the optical method of Near-Infrared-Spectroscopy (NIRS) is evaluated with regards to its capability of non-invasive detection of Leão´s cortical spreading depression (CSD) and spontaneous peri-infarct-depolarizations (PID). With the NIR-spectrometer NIRO 500 (Hamamatsu, Japan) regional cerebral oxy-genation (rCBO) changes were measured during CSD in 9, and during PID in 10 barbiturate anesthatized rats. The method if NIRS that relies on oxygen-dependent absorption changes of hemo-globin and cytochrome oxydase as well as the high penetrability of biologic tissues for light in the range between 700 and 1000 nm proved suitable to detect and to distinguish both CSD and PID experimentally. This distinction relies on the robust decrease of deoxy- and increase of oxyhemo-globin concentrations (i.e. a relative hyperoxemia) during CSD while PID is cha-racterized by an initial increase of deoxy- and decrease of oxyhemoglobin (relative hypooxemia). Despite the profound anatomical differences between gyrencephalic humans and lyssencephalic rats, the observed patterns of rCBO changes may guide the inter-pretation of future NIRS measurements in patients with migraines with aura (CSD) or stroke (PID). However, for concentration changes of oxydized cytochrome aa3 with its low con-centration compared to the hemoglobins, the pathophysiological interpretation of the data obtained with NIRO 500 is confounded by the limits of attenuation mea-surements at only four wavelengths. A validation of the cytochrome oxydase signal and an improved quantification of all concentration changes is highly desirable and may be achieved by employment of a continuous-wavelength device measuring the full spectral range of the near infrared. It would also allow to measure the mean optical pathlength in the highly scattering tissue and to correct for its physiologically occuring changes e.g. by measurements at the water absorption peak. Similar improvements would enhance the value of the method for further physiolo-gical and pathophysiological studies because NIRS provides the unique opportu-nity to obtain simultaneous data on blood oxygenation as well as the redox state of the mitochondrial cytochrome oxydase.

L’objectif de cette thèse est de démontrer le potentiel de la spectroscopie proche infrarouge multipoints pour suivre en ligne des milieux diffusants et absorbants dont les propriétés physique et chimique varient au cours du temps. Des travaux ont tout d’abord porté sur l’étude d’un milieu où seule la diffusion varie au cours du temps. Le suivi de la réaction de précipitation de la silice a été choisi pour illustrer un tel milieu. Dans un premier temps, des spectres en transmission collimatée ont été mesurés sur des échantillons prélevés au cours du procédé. Grâce à l’application de la loi de Beer-Lambert, il a été possible d'obtenir les coefficients de diffusion de chaque échantillon et ainsi retrouver les différentes étapes de la réaction. Une analyse en composante principale de ces coefficients a montré la présence de deux régimes de diffusion dans le milieu réactionnel. Dans un second temps, une sonde de mesures multipoints, a été immergée dans un réacteur de fabrication. Des spectres ont été mesurés à des angles de 30°, 90°, 150°, 170° et 180° par rapport à la source, tout au long de la précipitation. Une analyse univariée, aux différents angles, a montré des différences et des similitudes entre les positions, en lien avec les interactions lumière matière. Puis une analyse multivariée multi tableaux ACCPS (Analyses en Composantes Communes et Poids Spécifiques) a été appliquée. Elle permet de combiner les informations issues des différents angles et de montrer les informations communes et spécifiques. Cela a permis de valider l’utilisation de la mesure multipoints pour le contrôle du procédé. Grâce à l’analyse des scores globaux, différents types de diffusion et la spécificité de certains angles dans la détection de modifications physiques, ont été identifiés. L’analyse des loadings individuels, a confirmé les différentes étapes réactionnelles identifiées et a révélé des phénomènes relatifs à la diffusion de la lumière comme l’allongement du trajet optique. La même démarche, analyses en laboratoire puis en ligne, a été appliquée à un milieu où la diffusion et l’absorption varient. Le suivi des produits issus des procédés pour l’amélioration de l’extraction du pétrole a été choisi pour illustrer un tel milieu. Dans un premier temps, des microémulsions (eau, huile et tensio-actif) ont été fabriquées de sorte à ce que les propriétés chimiques et physiques soient différentes. Puis chaque phase a été analysée séparément. Une analyse en composantes principales a permis de caractériser les systèmes grâce à l’identification des phases. Elle a également montré que les différences entre elles étaient liées, à la fois à l’absorption et à la diffusion. Pour séparer ces deux phénomènes, une résolution multivariée de courbes par régression alternée a été appliquée. Cela a permis d’expliquer plus en détail les différences entre les phases et d’obtenir des suivis semi-quantitatifs. Enfin dans la dernière partie, les mêmes échantillons ont été analysés en circulation avec une sonde multipoints. Une ACCPS a de nouveau été appliquée et a montré que des phénomènes de diffusion différents avaient lieu entre les classes. De plus elle a mis en lumière une hétérogénéité dans la composition des phases, qui ne pouvait pas être détectée avec des mesures en statique. / The aim of this thesis is to demonstrate the potential of multipoint near infrared spectroscopy to monitor online scattering and absorbing environments, where physical and chemical properties change over time. The work was first based on the study of an environment where scattering was the only parameter varying over time. Monitoring of the silica precipitation reaction has been chosen to illustrate that environment. First of all, spectra in collimated transmittance mode were acquired from samples collected during the process. Thanks to the Beer-Lambert law, scattering coefficients of each sample were obtained and from them the different steps of the reaction were found. A principal component analysis of those coefficients showed the existence of two scattering modes into the reaction environment. In the second phase, a multipoint probe was submerged in an industrial reactor. Spectra were measured during the precipitation with angle of 30°, 90°, 150°, 170° and 180° to the source. A univariate analyse at those different angles showed differences and similarities between the positions, linked to the interaction of light and matter. Then, a multiblock and multivariate analysis, CCSWA (Common Component and Specific Weight Analysis), was applied. It allows the combination of information from different angles and also shows common and specific information, in order to validate the use of multipoint measurement to monitor the process. Thanks to the analysis of global scores, different kind of scattering and the specificity of some angle, in the detection of physical modifications were identified. The analysis of individual loadings confirmed the identification of the different steps of reaction and has revealed some phenomena related to light scattering, such as the extension of the optical path. The same approach, analysis in the laboratory then online, was applied on an environment where scattering and absorption vary. Monitoring of products from the process to improve oil extraction, has been chosen to illustrate that environment. First of all, micro emulsions (water, oil and surfactant) were made so that physical and chemical properties would be different between samples. Then, each phase was analysed separately. A principal component analysis was used to characterize the system by phase identification. It also showed the differences between the phases were linked, both to absorption and scattering. In order to separate those two phenomena, a multivariate resolution by alternate regression has been applied. It has permitted to explain in more details the differences between the phases and to obtain semi-quantitative monitoring. Finally, in the last part, the same samples were analysed in circulation with a multipoint probe. A CCSWA has been applied again and showed that different scattering phenomenon occurred between the categories. Moreover, it highlighted a heterogeneity in the phases' composition that couldn’t be detected in static measurement.

This project utilised a novel solvent-free continuous cocrystallisation (SFCC) method to manufacture pharmaceutical co-crystals. The objectives were to optimize the process towards achieving high co-crystal yields and to understand the behaviour of co-crystals under different conditions. Particular attention was paid to the development of near infrared (NIR) spectroscopy as a process analytical technology (PAT). Twin screw, hot melt extrusion was the base technique of the SFCC process. Changing parameters such as temperature, screw speed and screw geometry was important for improving the co-crystal yield. The level of mixing and shear was directly influenced by the screw geometry, whilst the screw speed was an important parameter for controlling the residence time of the material during hot melt extrusion. Ibuprofen – nicotinamide 1:1 cocrystals and carbamazepine – nicotinamide 1:1 co-crystals were successfully manufactured using the SFCC method. Characterisation techniques were important for this project, and NIR spectroscopy proved to be a convenient, accurate analytical technique for identifying the formation of co-crystals along the extruder barrel. Separate thermal and model shear deformation studies were also carried out to determine the effect of temperature and shear on co-crystal formation for several different pharmaceutical co-crystal pairs. Finally, NIR spectroscopy was used to create two partial least squares regression models, for predicting the 1:1 co-crystal yield of ibuprofen – nicotinamide and carbamazepine – nicotinamide, when in a powder mixture with the respective pure API. It is believed that the prediction models created in this project can be used to facilitate future in-line PAT studies of pharmaceutical co-crystals during different manufacturing processes.

Développer de nouveaux fluorophores ayant une forte section efficace d’absorption à deux photons (ADP) et des propriétés d'émission dans le rouge lointain est important, en particulier pour l’imagerie in-vivo profonde. Cette gamme de longueur d'onde correspond en effet à la fenêtre de transparence optique des tissus. Cette thèse étudie le potentiel de nouveaux fluorophores émettant dans rouge construits sur un noyau fluorène pour la microscopie à deux photons in-vivo, en privilégiant l'imagerie du système vasculaire, d'une part, et la mesure optique de la pression d'oxygène dissous, d'autre part.Ainsi, une famille de chromophores asymétriques a été conçue et synthétisée. La plupart des chromophores présentent une forte émission dans le proche IR, induite par l'agrégation. De plus, une stratégie de co-protection basée sur un système micellaire / silice a été utilisé pour préparer des nanoparticules avec un intérieur apolaire et conserver les propriétés optiques des chromophores dipolaires en solution aqueuse. Des mesures de fluorescence excitée à deux photons ont été menées en solvant organique et en suspension aqueuse. Les agrégats et les nanoparticules ont été utilisés avec succès en imagerie biphotonique du système vasculaire sur petit animal en utilisant un modèle de tumeur à l'intérieur de la peau de l'oreille de la souris. Les nanoparticules de silice montrent une coloration exceptionnelle du système vasculaire qui en fait de parfaits marqueurs du système vasculaire.Dans un deuxième temps, quatre nouveaux chromophores, absorbant à deux photons, ont été synthétisés et leurs propriétés photo physiques à un et à deux photons ont été étudiées. Le chromophore le plus adapté a ensuite été greffé de manière covalente par chimie click, à un complexe de palladium avec un ligand porphyrine, cœur phosphorescent dont l’émission est sensible à la présence d’oxygène. Deux composés contenant quatre ou huit absorbeur à deux photons ont été obtenus et étudiés. Les résultats démontrent que l'incorporation d'un chromophore ADP approprié peut effectivement augmenter les propriétés d’ADP du système, ce qui permet une sensibilité efficace vis-à-vis de l'oxygène. / The development of fluorophores with efficient two-photon absorption (TPA) and emission properties in the far red/NIR is important, especially for in depth in-vivo optical imaging as this wavelength range corresponds to the optical transparency window of tissues. This thesis investigates the potential of new red emitting fluorophores containing a fluorene ring for in-vivo two-photon microscopy focusing on vascular imaging on one hand and on oxygen pressure measurement on the other hand.A new series of asymmetrical fluorene-based chromophores were designed and synthesized. Their structure-property relationships were systematically investigated. It was found that most of chromophores exhibit aggregation-induced emission behaviors in the NIR region. In addition, a micelle/silica coprotection strategy was proposed to prepare nanoparticles with a less polar interior, which can be used to conserve optical properties of dipole chromophores in aqueous solution. The two-photon excited fluorescence (TPEF) measurements indicate that they all display obvious TPA activities in organic solvent and aqueous suspension. Both the NIR-emissive aggregates and nanoparticles have been successfully used for TPEF imaging of blood vessels inside mouse ear skin. The silica nanoparticles show outstanding staining of the vascular system making them perfect blood pool markers.On a second part, four new fluorene-based two-photon absorbing chromophores have been synthesized and their one- and two-photon photophysical properties were investigated. The optimum chromophore was successfully attached covalently to an oxygen responsive phosphorescent Pd-porphyrin complex by click chemistry. Two new compounds contain four or eight TPA chromophores donor connected to the phosphorescent core. The result demonstrate that the incorporation of a suitable TPA chromophore can effectively enhance the TPA of the system, allowing efficient sensitivity towards oxygen.

Made available in DSpace on 2014-12-17T15:41:52Z (GMT). No. of bitstreams: 1 MariaRCI_DISSERT.pdf: 3251511 bytes, checksum: ea24e2034f12435ffe63fbcaee9ff544 (MD5) Previous issue date: 2010-02-24 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / In this work we used chemometric tools to classify and quantify the protein content in samples of milk powder. We applied the NIR diffuse reflectance spectroscopy combined with multivariate techniques. First, we carried out an exploratory method of samples by principal component analysis (PCA), then the classification of independent modeling of class analogy (SIMCA). Thus it became possible to classify the samples that were grouped by similarities in their composition. Finally, the techniques of partial least squares regression (PLS) and principal components regression (PCR) allowed the quantification of protein content in samples of milk powder, compared with the Kjeldahl reference method. A total of 53 samples of milk powder sold in the metropolitan areas of Natal, Salvador and Rio de Janeiro were acquired for analysis, in which after pre-treatment data, there were four models, which were employed for classification and quantification of samples. The methods employed after being assessed and validated showed good performance, good accuracy and reliability of the results, showing that the NIR technique can be a non invasive technique, since it produces no waste and saves time in analyzing the samples / Neste trabalho foram utilizadas ferramentas quimiom?tricas para classificar e quantificar o teor de prote?nas em amostras de leite em p?. Empregou-se a espectroscopia NIR com reflect?ncia difusa associada a t?cnicas multivariadas. Primeiramente, realizou-se um m?todo explorat?rio das amostras atrav?s da an?lise das componentes principais (PCA),em seguida a classifica??o de modelagem independente para analogia de classes (SIMCA). Dessa forma se tornou poss?vel classificar as amostras que se agruparam por semelhan?as em sua composi??o. Por fim, as t?cnicas de regress?o por m?nimos quadrados parciais (PLS) e regress?o por componentes principais (PCR) permitiram a quantifica??o do teor de prote?nas nas amostras de leite em p?, comparadas com o m?todo de refer?ncia Kjeldahl. Um total de 53 amostras de leite em p? comercializadas nas regi?es metropolitanas de Natal, Salvador e Rio de Janeiro foram adquiridas para an?lise, em que ap?s pr?-tratamento dos dados obtidos, foram encontrados quatro modelos, os quais empregaram-se para a classifica??o e quantifica??o das amostras. Os m?todos empregados ap?s serem avaliados e validados apresentaram bom desempenho, demonstrando exatid?o e confiabilidade nos resultados obtidos, mostrando que a t?cnica NIR pode ser uma t?cnica n?o invasiva, uma vez que n?o produz res?duos e ainda economiza tempo na an?lise das amostras

Made available in DSpace on 2014-12-17T15:41:56Z (GMT). No. of bitstreams: 1 AlessandraMC_DISSERT.pdf: 3533058 bytes, checksum: de92ff683335d9aa6c119013acb5b5d0 (MD5) Previous issue date: 2011-08-15 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / In this work calibration models were constructed to determine the content of total lipids and moisture in powdered milk samples. For this, used the near-infrared spectroscopy by diffuse reflectance, combined with multivariate calibration. Initially, the spectral data were submitted to correction of multiplicative light scattering (MSC) and Savitzsky-Golay smoothing. Then, the samples were divided into subgroups by application of hierarchical clustering analysis of the classes (HCA) and Ward Linkage criterion. Thus, it became possible to build regression models by partial least squares (PLS) that allowed the calibration and prediction of the content total lipid and moisture, based on the values obtained by the reference methods of Soxhlet and 105 ? C, respectively . Therefore, conclude that the NIR had a good performance for the quantification of samples of powdered milk, mainly by minimizing the analysis time, not destruction of the samples and not waste. Prediction models for determination of total lipids correlated (R) of 0.9955, RMSEP of 0.8952, therefore the average error between the Soxhlet and NIR was ? 0.70%, while the model prediction to content moisture correlated (R) of 0.9184, RMSEP, 0.3778 and error of ? 0.76% / Neste trabalho foram constru?dos modelos de calibra??o para determinar os teores de lip?dios totais e umidade em amostras de leite em p?. Para isso, utilizou-se a espectroscopia no infravermelho pr?ximo por reflect?ncia difusa, aliado ? calibra??o multivariada. Inicialmente, os dados espectrais foram submetidos ? corre??o multiplicativa do espalhamento da luz (MSC) e alisamento de Savitzsky-Golay. Em seguida, as amostras foram divididas em subgrupos por aplica??o da an?lise por agrupamento hier?rquico das classes (HCA) e crit?rio de Ward Linkage. Desta forma, tornou-se poss?vel construir modelos de regress?o por m?nimos quadrados parciais (PLS) que permitiu a calibra??o e previs?o dos teores de lip?dios e umidade, com base nos valores obtidos por m?todos de refer?ncia de Soxhlet e secagem a 105 ? C, respectivamente. Portanto, conclui-se que o NIR apresentou um bom desempenho para quantifica??o de amostras de leite em p?, principalmente pela minimiza??o do tempo das an?lises, n?o destrui??o das amostras e n?o gera??o de res?duos. Os modelos de previs?o para determina??o de lip?dios totais apresentaram correla??o (R) de 0,9955, RMSEP de 0,8952, por conseguinte, o erro m?dio entre o Soxhlet e o NIR foi ? 0,70%, enquanto o modelo de previs?o para teor de umidade apresentou correla??o (R) de 0,9184, RMSEP, 0,3778 e erro de ? 0,76%

Made available in DSpace on 2014-12-17T15:42:05Z (GMT). No. of bitstreams: 1 AnaCON_DISSERT.pdf: 2560306 bytes, checksum: 01b4c26da4748508add3442c02e1b43b (MD5) Previous issue date: 2013-02-04 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior / In this work, the quantitative analysis of glucose, triglycerides and cholesterol (total and HDL) in both rat and human blood plasma was performed without any kind of pretreatment of samples, by using near infrared spectroscopy (NIR) combined with multivariate methods. For this purpose, different techniques and algorithms used to pre-process data, to select variables and to build multivariate regression models were compared between each other, such as partial least squares regression (PLS), non linear regression by artificial neural networks, interval partial least squares regression (iPLS), genetic algorithm (GA), successive projections algorithm (SPA), amongst others. Related to the determinations of rat blood plasma samples, the variables selection algorithms showed satisfactory results both for the correlation coefficients (R?) and for the values of root mean square error of prediction (RMSEP) for the three analytes, especially for triglycerides and cholesterol-HDL. The RMSEP values for glucose, triglycerides and cholesterol-HDL obtained through the best PLS model were 6.08, 16.07 e 2.03 mg dL-1, respectively. In the other case, for the determinations in human blood plasma, the predictions obtained by the PLS models provided unsatisfactory results with non linear tendency and presence of bias. Then, the ANN regression was applied as an alternative to PLS, considering its ability of modeling data from non linear systems. The root mean square error of monitoring (RMSEM) for glucose, triglycerides and total cholesterol, for the best ANN models, were 13.20, 10.31 e 12.35 mg dL-1, respectively. Statistical tests (F and t) suggest that NIR spectroscopy combined with multivariate regression methods (PLS and ANN) are capable to quantify the analytes (glucose, triglycerides and cholesterol) even when they are present in highly complex biological fluids, such as blood plasma / No presente trabalho, a an?lise quantitativa de glicose, triglicer?deos e colesterol (total e HDL) em plasma sangu?neo de ratos e humanos foi realizada sem necessidade de pr?tratamentos de amostras, atrav?s do uso da espectroscopia no infravermelho pr?ximo (NIR), aliada a m?todos multivariados. Para tanto, foram comparadas diferentes t?cnicas e algoritmos utilizados para pr?-processamentos de dados, sele??o de vari?veis e regress?es multivariadas, tais como a regress?o por m?nimos quadrados parciais (PLSR), regress?o n?o linear via redes neurais artificiais (ANN), regress?o por m?nimos quadrados parciais por intervalos (iPLS), algoritmo gen?tico (GA), algoritmo das proje??es sucessivas (SPA), entre outros. Para as determina??es em sangue de ratos, os algoritmos de sele??o de vari?veis apresentaram resultados satisfat?rios tanto em rela??o aos coeficientes de correla??o (R?) quanto para os valores de erro quadr?tico m?dio de previs?o (RMSEP) para os tr?s analitos, especialmente para triglicer?deos e colesterol-HDL. Os valores de RMSEP para glicose, triglicer?deos e colesterol-HDL atrav?s do melhor modelo PLS foram de 6,08, 16,07 e 2,03 mg dL-1, respectivamente. Para as determina??es em sangue de humanos, as previs?es atrav?s de modelos PLS apresentaram resultados insatisfat?rios, com comportamento n?o linear e presen?a de bias. A regress?o ANN foi investigada como uma alternativa ao PLS, por sua habilidade de modelar sistemas n?o lineares. O erro quadr?tico m?dio de monitoramento (RMSEM) para glicose, triglicer?deos e colesterol total, para os melhores modelos ANN, foram de 13,20, 10,31 e 12,35 mg dL-1, respectivamente. Testes estat?sticos (F e t) sugerem que a espectroscopia NIR aliada a m?todos de regress?o multivariados (PLS e ANN) possuem capacidade de quantificar os analitos (glicose, triglicer?deos e colesterol) mesmo quando os mesmos est?o presentes em fluidos biol?gicos altamente complexos, como o plasma sangu?neo

As Boas Práticas de Fabricação de Medicamentos (BPFM) enfatizam que a indústria farmacêutica deve dirigir seus esforços no sentido de compreender a variação do processo, incluindo as fontes, o grau de variação e o impacto dessa variação nas características de qualidade do produto. O processo de fabricação de medicamentos tem apresentado significativas mudanças, em especial no que se refere à introdução de tecnologias analíticas que permitem o controle do processo em tempo real. A abordagem baseada na análise de risco e no novo Sistema de Qualidade Farmacêutica constitui ponto central das BPFM para o século XXI. Os órgãos regulatórios têm exigido da indústria farmacêutica sua adesão na melhoria contínua relativa ao desempenho de seus processos e, por consequência, na qualidade do produto. O objetivo do presente trabalho foi o desenvolvimento e validação de método analítico empregando espectroscopia no infravermelho próximo, assim como a avaliação do processo de fabricação de comprimidos de Captopril 25 mg, empregando abordagem racional-científica. Com referência à avaliação do processo, foram adotadas as seguintes ferramentas: análise de modos e efeitos de falhas (FMEA); gráficos de controle; índices de capacidade e análise de variância (ANOVA). A espectroscopia por infravermelho próximo (NIR) foi selecionada por apresentar maior rapidez na obtenção dos resultados, maior simplicidade na preparação das amostras, multiplicidade das análises a partir de uma única leitura e por apresentar característica não invasiva. Os resultados comprovaram a adequação dessa tecnologia na avaliação quantitativa do Captopril nas etapas de mistura de pós e de compressão. Os desvios padrão relativos na determinação da uniformidade de Captopril na mistura de pós e nos comprimidos empregando método no NIR foram, respectivamente 3,15 e 0,18%. No que se refere à avaliação da estabilidade e da capacidade do processo, as ferramentas adotadas permitiram a compreensão das fontes de variabilidade, assim como a determinação de seu grau, nas diferentes etapas do processo. Os índices de capacidade (CpK) relativos à uniformidade de Captopril (% p/v) na mistura de pós, ao peso médio do comprimido, à uniformidade de conteúdo e à % (p/v) dissolvida de Captopril, no ensaio de dissolução, foram 0,70, 1,94, 1,80 e 2,19, respectivamente. / The Good Manufacturing Practices (GMP) for Medicinal Products point out that the pharmaceutical industry must direct efforts to understand the variation of the processes, including the sources, the level of variation and the variation impact on the process in characteristics of the product. The manufacturing process has shown meaningful changes, especially in the introduction of new analytical technologies that allow the process control in real time. The approach based on risk analyses and on the new Pharmaceutical Quality System is a central key for the GMP for the XXI century. The Regulatory Agencies have demanded the pharmaceutical industry to adhere the continuous improvement related to the performance of its processes and, consequently, the product quality. Thus, the present paper aimed the development and validation of the analytical method employing NIR spectroscopy as the assessment of manufacturing process of Captopril 25 mg tablets, using rational scientific approach. Regarding the process assessment, the following tools were adopted: analysis of failure modes and effects analysis (FMEA), control charts, capability indexes, as well as analysis of variance (ANOVA). The near-infrared spectroscopy was selected due to its greater speed in getting the results, simplicity in sample preparation, and multiplicity of analysis from a single reading and provide non-invasive feature. The results confirmed the suitability of this technology in quantitative assessment of Captopril on the steps of mixing powders and compression. The relative standard deviations for the determination of Captopril uniformity in the post mixtures and in the tablets employing NIR were 3,15 e 0,18%, respectively. In reference to the stability assessment and process capacity, the tools adopted permitted the understanding of the sources of variability, as well as the determination of their level in different phases of the process. The capacity indexes relating to Captopril uniformity (% p/v) in the powder mixture, the average weight of the tablet, the content uniformity and the % (p/v) dissolved Captopril, in the dissolution assay were 0,70, 1,94, 1,80 and 2,19, respectively.