The present work aimed at assessing the dose emission and aerodynamic particle size characteristics of formoterol fumarate from Atimos Modullite, a metered dose inhaler (MDI) and Foradil Aeroliser, Easyhaler, and Oxis Turbuhaler dry powder inhalers (DPI) at different inhalation flow rates and volumes using in vitro methodology. Recognised methods have been adopted and validated to generate the results. The in vitro characteristics of formoterol were measured according to standard pharmacopeial methodology with adaptation to simulate routine patient use. The dose emission from the Atimos Modulite was determined using inhalation volumes of 4 and 2 L and inhalation flows of 10, 28.3, 60, and 90 L/min. The %nominal dose emitted was consistent between the various flow rates and inhalation volumes of 4 and 2L. The particle size distribution was measured using an Anderson Cascade Impactor (ACI) combined with a mixing inlet valve to measure particle size distribution at inhalation flow rates below 30 L/min. The particle size distribution of formoterol from Atimos Modulite was measured using inhalation flows of 15, 28.3, 50, and 60 L/min with and without different spacers, Aerochamber and Volumatic. The mean fine particle dose (%nominal dose) through an Atimos without spacer were 53.52% (2.51), 54.1% (0.79), 53.37% (0.81), 50.43% (1.92) compared to Aerochamber 63.62% (0.44), 63.86% (0.72), 64.72% (0.47), 59.96% (1.97) and Volumatic 62.40% (0.28),63.41% (0.52), 64.71% (0.61), 58.43% (0.73), respectively. A small decrease in the fine particle dose was observed as the inhalation flow increased, but this was not significant. The respective mean mass aerodynamic diameter (MMAD) increased as the flow rate was increased from 15 of 60 L/min. Results also suggests that the use of spacers provides better lung deposition for patients with problems using MDI. The dose emission from the Foradil Aeroliser was determined using inhalation volumes of 4 and 2 L, at inhalation flows of 10, 15, 20, 28.3, 60, and 90 L/min plus two inhalations per single dose. The %nominal dose emitted using 2 L inhalation volume was approximately half when compared to results obtained using inhalation volume of 4 L. A significantly (p<0.001) higher amount of drug was also emitted from Easyhaler® at inhalation volume of 4 L through flow rates of 10, 20, 28.3, 40, and 60 L/min compared 2 L. Similar results were observed through Oxis Turbuhaler at inhalation flow rates of 10, 20, 28.3, 40, and 60 L/min. Comparative studies were also carried out to evaluate the particle size distribution of formoterol through the DPIs. The nominal fine particle dose through Aeroliser using inhalation flows of 10, 20, 28.3, 60 and 90 L/min were 9.23%, 14.70 %, 21.37%, 28.93%, and 39.70% for the 4 L and 4.17%, 5.55%, 7.28%, 8.41%, and 11.08% for the 2 L, respectively. The respective MMAD significantly (p<0.001) decreased with increasing flow rates. Aeroliser performance showed significant (p<0.001) increase in the % nominal fine particle dose for two inhalations compared to one inhalation at both 4 and 2 L. The Easyhaler was measured using inhalation flows of 10, 20, 28.3, 40, 60 L/min. The nominal fine particle dose were 19.03%, 27.09%, 36.89%, 49.71% and 49.25% for the 4 L and 9.14%, 15.44%, 21.02%, 29.41%, 29.14% for the 2 L, respectively. The respective MMAD significantly (p<0.001) decreased with increasing flow rates. Easyhaler performance at both 4 and 2 L showed no significant differences between one and two inhalations at low flow rates (10, 20, 28.3), but this was significant (p<0.05) at higher flow rates (40 and 60 L/min). The Oxis Turbuhaler was also measured using inhalation flows of 10, 20, 28.3, 40, 60 L/min. The nominal fine particle dose were 12.87%, 24.51%, 28.25%, 34.61%, 40.53% for the 4 L and 8.55%, 15.31%, 21.36%, 19.53%, 22.31% for the 2 L, respectively. Turbuhaler performance showed significant (p<0.05) differences between one and two inhalations at varying flow rates 2 L inhalation volumes, but not at 4 L. The use of Foradil Aeroliser delivers small particles as the Oxis Turbuhaler using two inhalations hence delivering formoterol deep into the lungs. Also, this thesis shows that high flow resistance of Turbuhaler will indeed influence the ability of patients with severe asthma or children to use the system. Beside, Easyhaler produced the highest drug delivery to the lungs, thus, making it a more desirable system to use, especially for children and asthma sufferers.

External beam radiotherapy (RT) is the primary treatment modality for patients with inoperable lung tumours. Respiration-induced motion and related intra-/interfractional variations present a series of limitations to the success of existing conventional treatment modalities for lung cancer. Subsequently, to minimise the effects of respiration different management techniques have been proposed and are available. Respiratory gated radiotherapy (RGRT) holds promise to improve dose conformity, reduce the normal tissue control probability while increasing the tumour control probability. Its effectiveness depends on precise tumour localisation and targeting during dose delivery. In this thesis, the suitability of RGRT for the compensation of breathing induced motion was investigated by means of phantom studies and film dosimetry. Both regular and irregular trajectories were simulated during gated dose delivery and their effects on dose distributions analysed. Respiration-induced motion led to dose blurring and hence to less conformal dose distributions, which resulted overall in underdose of the treatment planning volume and an overdose of healthy surrounding tissue. Compared to non-gated dose delivery, RGRT improved dose conformity by enabling steeper dose gradients, resulting in an increased sparing of healthy tissue, at the expenses of increased delivery times. In the presence of irregular motion paths the dosimetric advantages of RGRT were observed to decrease. In the absence of a clinical tool for treatment verification such irregularities may pass unnoticeable and may lead to poor treatment outcomes. Investigations of the suitability of a software tool for tracking lung tumours in portal images during RGRT demonstrated that it is possible to determine and track tumour motion during gated treatment. Both the residual tumour motion inside the gating window as well as the probability density function were used as measures to quantify tumour position and variability. Tracking information was sufficient to quantify residual motion and variability. Baseline drifts as well as sudden fluctuations in tumour positions were detected and quantified, which led to considerable variations in residual motion which in turn may result in marginal miss. Although this was a retrospective analysis of motion data, the tool showed a great potential for verification of the tumour position during RGRT and may possibly be useful for adaptation of the gating window.

Studies on relations between number of sessions and effect of psychotherapy have usually assumed a constant rate of change across different lengths of therapy, explained by a model called the dose-effect model. This assumption has been challenged by the good-enough level (GEL) model, which makes the prediction that the rate of change will vary as a function of total number of sessions. This study aimed to compare these models. We also assessed the relationship between reliable and clinically significant change (RCSI) and total dose of therapy. Participants were drawn from two datasets in the Swedish primary care (n = 640) and adult psychiatric care (n = 249). The participants made session-wise ratings on the Clinical Outcomes in Routine Evaluation-Outcome Measure (CORE-OM). Multilevel analyses indicated a better fit using the GEL-model, with some reservations concerning RCSI and patterns of change. The results may indicate a general lawful relationship that may have implications for future research, as well as psychotherapy practice and policy making.

Made available in DSpace on 2014-10-09T12:44:19Z (GMT). No. of bitstreams: 0 / Made available in DSpace on 2014-10-09T13:57:02Z (GMT). No. of bitstreams: 1 06870.pdf: 3530366 bytes, checksum: 5bddd2a5ab0508e81a57fdbb2c5f73d1 (MD5) / Dissertacao (Mestrado) / IPEN/D / Instituto de Pesquisas Energeticas e Nucleares - IPEN/CNEN-SP

Orientador: Marco Antônio Rodrigues Fernandes / Resumo: Este trabalho apresenta um estudo de planos radioterápicos realizados com o Sistema de Planejamento de Tratamento (TPS) XiO 5.10, utilizados em centros de radioterapia. Foram analisados os protocolos clínicos de radioterapia através de pesquisa bibliográfica em artigos científicos e textos produzidos pelas sociedades especializadas da área da radioncologia. Foram realizadas visitas técnicas junto a serviços de radioterapia para melhor entendimento das diferenças entre os distintos sistemas de planejamentos radioterápicos aplicados. A pesquisa verificou a resposta de três diferentes algoritmos computacionais de cálculo de dose de radiação: Superposition, Convolution e Fast Superposition, disponíveis no TPS XiO® da Elekta Medical Systems. Para isto foram analisados os parâmetros de distribuição de dose para 22 planejamentos radioterápicos realizados em uma clínica de radioterapia privada, sendo: 1 paciente com metástase cerebral, 3 pacientes com tumores de cabeça e pescoço, 9 com câncer de mama e 9 com tumores da próstata. Também foi quantificada a influência da heterogeneidade do tecido irradiado, através dos métodos de correção de heterogeneidade do XiO. Independentemente da correção de heterogeneidade do tecido, de um modo geral, o algoritmo de Convolution tendeu a subestimar a dose no PTV quando comparado com os outros dois algoritmos de cálculo de dose. Não foram percebidas diferenças significativas nos valores de dose de radiação calculados para o PTV, determinadas com os... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: This paper bring forward a study of radiotherapeutic planning accomplished by XiO 5.10 treatment planning systems utilized in radiotherapic centers. The clinical protocols of radiotherapy were analyzed through bibliographic research in scientific articles and texts produced by societies related to the area of radioncology. . Technical reconnaissance visits were performed in radiotherapy services for a better understanding of the differences between the systems of radiotherapy planning applied. The research verified the response of three distinct computational algorithms of calculation of dose radiation: Superposition, Convolution and Fast Superposition, available in the Tps XiO® of Elekta Medical Systems. For this purpose, were analyzed the dose distribution parameters for 22 radiotherapy plans performed in a private radiotherapy clinic: 1 patient with cerebral metastasis, 3 patients with head and neck tumors, 9 with breast cancer and 9 with prostate tumors . The influence of the heterogeneity of the irradiated tissue was also quantified through the XiO heterogeneity correction methods. Regardless of the correction of tissue heterogeneity, the Convolution algorithm tended to, in general, underestimate the dose in the PTV when compared to the other two dose calculation algorithms. No significant differences were observed in the radiation dose values calculated for the PTV, determined with the other two algorithms, however, the Superposition method provides a slight reduction o... (Complete abstract click electronic access below) / Mestre

Le développement dans l'industrie aéronautique des composites à matrices organiques renforcés par des fibres de carbone se justifie entre autres par leur très bonne capacité à transférer les efforts mécaniques élevés en rapport de leur faible masse. Les matrices Poly-EtherEtherCétone (PEEK) sont des candidates idéales pour les applications structurales en raison de leurs bonnes propriétés chimiques, mécaniques, ainsi que d'une stabilité thermique supérieure à d'autres polymères. Cependant celles-ci présentent des inconvénients majeurs pour la phase de mise en peinture, à savoir une faible réactivité de surface et une inertie chimique importante qui impliquent le développement d'un procédé d'activation de surface avant toute étape de revêtement. Ce travail se concentre sur une technologie de plasma froid à la pression atmosphérique développée par la société AcXys Technologie(r) : le module ULS en post-décharge. Cette technologie, choisie sur la base de considérations industrielles et environnementales, vise à garantir une activation de surface efficace, sans détérioration des propriétés intrinsèques de la matrice polymère. L'objectif de ce travail est alors d'améliorer la compréhension des mécanismes d'activation par plasma menant à une meilleure adhésion à l'interface composite PEEK/revêtement industriel. Des caractérisations électriques et optiques de la décharge et de la post-décharge nous permettront de mieux comprendre le fonctionnement intrinsèque de l'équipement. Tandis que la caractérisation des modifications de surface (de nature chimiques, physiques et/ou physico-chimiques), notamment au travers de la mouillabilité et de l'acido-basicité, permet d'appréhender l'influence des paramètres opératoires et du gaz plasmagène utilisé. Dans la perspective de mieux spécifier les conditions optimales d'utilisation industrielles, nous évaluons dans cette étude la pertinence de l'utilisation de la notion de dose plasma pour ce procédé, notion représentative de l'ensemble des paramètres opératoires. Cette notion, communément rencontrée dans d'autres procédés, est revue et développée pour correspondre à notre cas d'étude. Son utilisation permet une juste comparaison entre des configurations d'activation différentes puisqu'elle rend compte à la fois de la puissance dissipée sur le substrat et du temps d'interaction plasma/surface. Afin de valider la dose plasma exprimée pour notre équipement, nous introduisons des mesures d'adhérence effectuées entre le composite activé et le revêtement. Les résultats laissent entrevoir des perspectives d'améliorations dans les procédés industriels d'activation de surface. / Carbon Fiber Reinforced Polymer (CFRP) using thermoplastic polymer matrices as Poly-EtherEtherKetone (PEEK) for example, are increasingly being used in structural engineering due to their light weight coupled with good mechanical properties. In aeronautic industry, the adhesive bonding of these composites is often required. However, their low surface energy motivates the development of robust and reliable surface activation treatments aiming at increasing the surface reactivity before painting. For this reason, we have used an atmospheric pressure plasma torch developed by AcXys Technologies(r). It is a remote plasma, well known to be an effective process to improve surface reactivity without deterioration of the bulk matrix properties. This easily implementable technology attracts many industries looking for a cost-effective and eco-friendly surface activation process. The aim of this work is to contribute to the understanding of plasma activation mechanisms leading to a greater and a more durable adhesion between PEEK matrix and an industrial painting. This study provides some insight into the effects of process parameters (device power, distance between nozzle and substrate, etc) on adhesion improvement. Mechanisms which are attributed to it are investigated by means of three point bending and crosscut adhesion standard tests. Results are discussed with respect to surface properties characterized by wettability measurements including acid-base approach, X-ray Photoelectrons Spectroscopy and Atomic Force Microscopy. In order to facilitate industrialization of atmospheric pressure remote plasma as surface activation technique, this study proposes a new approach aiming at allowing a better and more equitable comparison between atmospheric pressure plasma processes. This comparison is made through the "plasma dose" expression, similar to the one commonly used in Corona process. It is applicable for one plasma gas and accounts for both received energy and interaction time of the post-discharge with the surface. Mechanisms assigned to improvement of adhesion will be more appreciated and their characterization will contribute to a greater definition of industrial surface preparation range by remote plasma.

Nesta dissertação é apresentado o estudo da célula SRAM estática de 6 transistores, com tecnologia CMOS, sendo utilizada em ambiente exposto à radiação. Foi verificado, através de simulação com o Hspice (HSPICE, 2009; KIME, 1998) e com a análise de Monte Carlo, o seu comportamento com relação à dose de ionização total (Total Ionization Dose, TID), a qual altera a tensão de limiar (threshold voltage, Vth) e a corrente de fuga, não sendo utilizada nenhuma técnica de fabricação especial para tolerância à radiação. Na simulação foi observado o comportamento da célula com relação ao tempo de atraso de escrita, à margem de ruído de leitura e ao consumo de energia. As simulações incluem as tecnologias de 130nm e 350nm sendo, portanto, possível comparar os efeitos de radiação citados em ambas, para verificar qual é a mais naturalmente resistente a radiação, verificando se está coerente com resultados divulgados na literatura. Para simular o efeito de dose, altera-se a tensão de limiar (threshold voltage, Vth) com a análise de Monte Carlo e, para a corrente de fuga, adiciona-se uma fonte de corrente entre o dreno e fonte de cada transistor. Os valores de Vth e corrente de fuga foram obtidos nas referências (HAUGERUD, 2005) para a tecnologia 130nm e (LACOE, 1998) para a tecnologia 350 nm. As simulações mostram que o comportamento foi coerente com resultados já conhecidos, em que a tecnologia mais antiga (350nm) tem alterações mais significativas do que a tecnologia mais atual, em relação à TID. / This work presents the study of the static RAM (SRAM) cell with 6 transistor, using CMOS technology, under radiation environment. The electrical behavior of the cell is evaluated using SPICE simulation (HSPICE, 2009; KIME, 1998) and applying Monte Carlo analysis. The effect of total ionization dose is analyzed through the modeling of threshold voltage shifts and leakage currents. The case study processes of this work do not use any special fabrication steps to make the circuit tolerant to radiation. The behavior of the cell related to write propagation time, read noise margin and energy consumption is evaluated through scripts written to support the simulation campaign. The simulations were performed for both 130nm and 350nm technologies, making possible to compare which one is more resistant to radiation. To further explore the dose effect in the case where the radiation does not affect all transistors in exactly the same way, the threshold voltage (Vth) of the transistors is varied randomly in the Monte Carlo analysis. To consider the leakage current, it is added a current source between drain and source of each transistor. The values of Vth and leakage current were obtained in reference (HAUGERUD, 2005) for the 130nm and in reference (LACOE, 1998) for the 350nm technology. The simulations show that the behavior was consistent with results already known, in which the older technology (350nm) is more significant changes then the most current technology, for the TID.

Les nouvelles techniques de la radiothérapie (Stéréotaxie, IMRT, VMAT, IGRT...etc) utilisent des faisceaux de photons de très petite taille (mini-faisceaux) dans le cas de petite tumeur, au cerveau par exemple, afin d'irradier précisément la lésion. En effet, leur taille de champ est inférieure à 3cm×3cm à 100 cm de la source de rayonnement, cependant la mesure de la dose dans les mini-faisceaux est caractérisée par de forts gradients de dose et un manque d'équilibre électronique latéral, nécessitant l'utilisation de détecteurs ayant un volume sensible et une résolution spatiale adaptés, avec une équivalence-eau aussi bonne que possible afin d'améliorer la précision de la dose mesurée. Les détecteurs commercialisés ne remplissent parfaitement ces conditions. Actuellement, il n'existe pas de consensus méthodologique international, ni de référence métrologique pour mesurer la dose dans les mini-faisceaux. Le protocole IAEA 398 utilisé pour calculer la dose absorbée dans un faisceau de 10×10 cm², n'est plus approprié pour les mini-faisceaux. Ce travail compare la mesure des données dosimétriques par différents détecteurs conçus pour ce type de faisceau et optimise celui le plus proche de la réalité. En absconse de référence métrologique, la vérification de l'ensemble de la mesure des données dosimétriques est assurée par l'utilisation des films gafchromiques du fait de son excellente résolution spatiale. Cette étude propose une méthode expérimentale pour estimer la dose délivrée en stéréotaxie intracrânienne. Cette méthode est basée sur la mesure de la dose de fuite en un point situé en dehors du champ d'irradiation. / The advanced techniques of radiotherapy use very small fields in case small tumors such as in the brain to irradiate precisely the lesion. This work concerns the measurement absorbed dose in small field of 0.5×0.5cm² to 3×3cm². However, the measurement dose in small fields is characterized by high gradient dose and a leak of lateral electronic equilibrium. That requires use a detector having an adapted sensitive volume and adapted spatial resolution. The detectors marketed are not perfectly compatible with these conditions. Actually, there is no international methodological consensus, nor a metrological reference for measurement dose in small fields. The IAEA (International Atomic Energy Agency) protocol 398 used to calculate the absorbed dose at 10cm×10cm isn't suitable for small fields. In absence a referenced detector, the dosimetric data measurement is verified using a Gafcromic films due to its excellent spatial resolution. We measure using conventional detectors (ionization chambers and/or Gafcromic film) the leakage dose at a point outside of irradiated field. The dosimetric data such as output factor OF, depth PDD percentage depth dose and dose profile OAR were also carried out by the diode. The correlation between the on-axis dose and off-axis dose is the subject of our study. This study proposes an experimental method to calculate the on-axis dose in small field for stereotactic radiotherapy. The method is based on the out of field leakage measurement. This model can be used to validate dose and output factor measurement. The experimental validation of the present method was performed for square and rectangular fields with sizes ranging from 0.5cm×0.5cm to 10cm×10cm.

Nesta dissertação é apresentado o estudo da célula SRAM estática de 6 transistores, com tecnologia CMOS, sendo utilizada em ambiente exposto à radiação. Foi verificado, através de simulação com o Hspice (HSPICE, 2009; KIME, 1998) e com a análise de Monte Carlo, o seu comportamento com relação à dose de ionização total (Total Ionization Dose, TID), a qual altera a tensão de limiar (threshold voltage, Vth) e a corrente de fuga, não sendo utilizada nenhuma técnica de fabricação especial para tolerância à radiação. Na simulação foi observado o comportamento da célula com relação ao tempo de atraso de escrita, à margem de ruído de leitura e ao consumo de energia. As simulações incluem as tecnologias de 130nm e 350nm sendo, portanto, possível comparar os efeitos de radiação citados em ambas, para verificar qual é a mais naturalmente resistente a radiação, verificando se está coerente com resultados divulgados na literatura. Para simular o efeito de dose, altera-se a tensão de limiar (threshold voltage, Vth) com a análise de Monte Carlo e, para a corrente de fuga, adiciona-se uma fonte de corrente entre o dreno e fonte de cada transistor. Os valores de Vth e corrente de fuga foram obtidos nas referências (HAUGERUD, 2005) para a tecnologia 130nm e (LACOE, 1998) para a tecnologia 350 nm. As simulações mostram que o comportamento foi coerente com resultados já conhecidos, em que a tecnologia mais antiga (350nm) tem alterações mais significativas do que a tecnologia mais atual, em relação à TID. / This work presents the study of the static RAM (SRAM) cell with 6 transistor, using CMOS technology, under radiation environment. The electrical behavior of the cell is evaluated using SPICE simulation (HSPICE, 2009; KIME, 1998) and applying Monte Carlo analysis. The effect of total ionization dose is analyzed through the modeling of threshold voltage shifts and leakage currents. The case study processes of this work do not use any special fabrication steps to make the circuit tolerant to radiation. The behavior of the cell related to write propagation time, read noise margin and energy consumption is evaluated through scripts written to support the simulation campaign. The simulations were performed for both 130nm and 350nm technologies, making possible to compare which one is more resistant to radiation. To further explore the dose effect in the case where the radiation does not affect all transistors in exactly the same way, the threshold voltage (Vth) of the transistors is varied randomly in the Monte Carlo analysis. To consider the leakage current, it is added a current source between drain and source of each transistor. The values of Vth and leakage current were obtained in reference (HAUGERUD, 2005) for the 130nm and in reference (LACOE, 1998) for the 350nm technology. The simulations show that the behavior was consistent with results already known, in which the older technology (350nm) is more significant changes then the most current technology, for the TID.

Made available in DSpace on 2014-06-11T19:28:23Z (GMT). No. of bitstreams: 0 Previous issue date: 2010-02-24Bitstream added on 2014-06-13T18:57:29Z : No. of bitstreams: 1 souza_fa_me_jabo.pdf: 1931388 bytes, checksum: 4c66f4cd56386572eb7ad4b52f5d1472 (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este trabalho avaliou a influência da heterocedasticidade e dos níveis nutricionais (número e posição) utilizados em ensaios dose-resposta, na estimativa do nível-ótimo e no ajuste dos modelos, além de verificar o quão informativas são as estatísticas utilizadas para avaliar a precisão do ajuste (R², R² ajustado, CV e SQD). Utilizaram-se dados dos experimentos realizados por Nascimento et al. (2007) e Siqueira (2009) e dados simulados. Constatou-se que, quando os níveis estiveram distribuídos próximos do verdadeiro requerimento, os modelos com platô proporcionaram resultados mais confiáveis. Já os modelos quadrático e exponencial se mostraram mais adequados para situações no qual os níveis estão mais dispersos em relação ao verdadeiro requerimento. A heterocedasticidade não interferiu na estimativa do nível-ótimo, porém influenciou no ajuste dos modelos e proporcionou pequenas mudanças nos parâmetros das equações obtidas. O coeficiente de determinação (ajustado e não ajustado) foi diretamente influenciado pela definição do nível mais próximo do ótimo e dos níveis extremos, enquanto que, o coeficiente de variação e a soma dos quadrados dos desvios, pelos níveis iniciais e pelo nível próximo do ótimo. A soma dos quadrados dos desvios demonstrou ser mais sensível, pois seu valor apresentou pequenas variações entre os modelos nas diferentes situações, que as outras estatísticas não detectaram. Ressalta-se a importância de se estabelecer corretamente o intervalo dos níveis estudados para que a dispersão dos valores do nível-ótimo estimado seja minimizada e que o ajuste seja satisfatório, independente do modelo utilizado / This work evaluated the influence of heteroskedasticity and the nutritional levels (number and position) used in dose-response trials to estimating the optimal-level and the adjustment of the models, also check how informative are the statistics used to evaluate the accuracy of fit (R ², R ² adjusted, CV and SQD). The data used in this experiment are from Nascimento et al. (2007) and’ Siqueira (2009) trials and simulated data. It was found that when levels were distributed close to the real requirement, the models with plateau have provided more reliable results. Since the quadratic and exponential models were more suitable for situations in which the levels are more dispersed about the real requirement. The heteroskedasticity did not affect the estimate of the level-optimal, but influenced the adjustment of the models and provided small changes in the parameters of the equations obtained. The coefficient of determination (adjusted and unadjusted) was directly influenced by the definition of the level closest to the optimum and extreme levels, while the coefficient of variation and the sum of squares of deviations were influenced by the initial levels and the level close to the optimum. The sum of squares of deviations was more sensitive, because its value showed small variations between models in different situations that the other statistics did not detect. Emphasized the importance to precisely define the range of levels studied to the dispersion of the obtained optimal-level is minimized and the fit is satisfactory regardless of the model