Cervical Cancer is the second most common type of cancer in women worldwide (500,000 cases/year) and one of the leading causes of cancer-related mortality in women in developing countries (230,000 cases/year). The Spectrx LightTouch™ device uses light to detect chemical and structural changes in cervical tissue. Light responds differently when exposed to normal cells and cancerous cells. The purpose of this research is to find the best model that can be used to diagnose the early cervical cancerous conditions. To achieve this goal, we first tried to reduce the number of variables. We use statistical and non-statistical methods to search for useful explanatory variables. Partial Least Square, Logistic Regression, CART, MARS, SVM have been used to build models. Bootstrap was adopted to estimate the threshold of PLS model. Comparison of the results indicates that PLS produces relatively better model in terms of the performances and to control over model threshold.

This multidisciplinary thesis is concerned with the prediction of drug formulations for the reduction of drug toxicity. Both scientific and computational approaches are utilised to make original contributions to the field of predictive toxicology. The first part of this thesis provides a detailed scientific discussion on all aspects of drug formulation and toxicity. Discussions are focused around the principal mechanisms of drug toxicity and how drug toxicity is studied and reported in the literature. Furthermore, a review of the current technologies available for formulating drugs for toxicity reduction is provided. Examples of studies reported in the literature that have used these technologies to reduce drug toxicity are also reported. The thesis also provides an overview of the computational approaches currently employed in the field of in silico predictive toxicology. This overview focuses on the machine learning approaches used to build predictive QSAR classification models, with examples discovered from the literature provided. Two methodologies have been developed as part of the main work of this thesis. The first is focused on use of directed bipartite graphs and Venn diagrams for the visualisation and extraction of drug-vehicle relationships from large un-curated datasets which show changes in the patterns of toxicity. These relationships can be rapidly extracted and visualised using the methodology proposed in chapter 4. The second methodology proposed, involves mining large datasets for the extraction of drug-vehicle toxicity data. The methodology uses an area-under-the-curve principle to make pairwise comparisons of vehicles which are classified according to the toxicity protection they offer, from which predictive classification models based on random forests and decisions trees are built. The results of this methodology are reported in chapter 6.

This multidisciplinary thesis is concerned with the prediction of drug formulations for the reduction of drug toxicity. Both scientific and computational approaches are utilised to make original contributions to the field of predictive toxicology. The first part of this thesis provides a detailed scientific discussion on all aspects of drug formulation and toxicity. Discussions are focused around the principal mechanisms of drug toxicity and how drug toxicity is studied and reported in the literature. Furthermore, a review of the current technologies available for formulating drugs for toxicity reduction is provided. Examples of studies reported in the literature that have used these technologies to reduce drug toxicity are also reported. The thesis also provides an overview of the computational approaches currently employed in the field of in silico predictive toxicology. This overview focuses on the machine learning approaches used to build predictive QSAR classification models, with examples discovered from the literature provided. Two methodologies have been developed as part of the main work of this thesis. The first is focused on use of directed bipartite graphs and Venn diagrams for the visualisation and extraction of drug-vehicle relationships from large un-curated datasets which show changes in the patterns of toxicity. These relationships can be rapidly extracted and visualised using the methodology proposed in chapter 4. The second methodology proposed, involves mining large datasets for the extraction of drug-vehicle toxicity data. The methodology uses an area-under-the-curve principle to make pairwise comparisons of vehicles which are classified according to the toxicity protection they offer, from which predictive classification models based on random forests and decisions trees are built. The results of this methodology are reported in chapter 6.

Foi realizado um estudo prospectivo em dez crianças (média de idade de apresentação da doença, 3.0 ± 1.6 anos) usando ciclosporina (CSA) uma microemulsão utilizada no tratamento da síndrome nefrótica idiopática (SNI) - com função renal normal avaliada pelo clearance de creatinina estimado pela estatura, que apresentavam o C0 (nível do vale) entre 50 e 150 ng/ml e que encontraram remissão completa da doença com a CSA. O objetivo foi comparar os parâmetros farmacocinéticos da CSA entre crianças com SNI em remissão ou descompensadas da doença. O perfil farmacocinético da CSA foi avaliado através da área sob a curva de 12 horas (AUC012) utilizando sete pontos. Esse procedimento foi realizado no mesmo paciente durante a remissão e a descompensação da doença com a mesma dose de CSA em mg/kg. A AUC012 foi calculada através de régua trapezoidal. Todos os pontos da curva e a AUC04 simplificada foram comparados com a AUC012. Não foi encontrada diferença significativa em quaisquer pontos da curva nos dois estados da doença, nem mesmo quando normalizados pela dose em mg/kg. Nosso estudo mostrou que a AUC04 foi o principal parâmetro em ambos os estados da doença (remissão e recidiva) quando comparado com a AUC012 (r=0.95 na remissão e r=0.93 na recidiva), e o segundo melhor parâmetro foi o C2 (r=0.86 na remissão e r=0.80 na recidiva). Mais estudos controlados são necessários para reproduzir melhor estes achados e confirmar nossa proposta. Além disso, o C2 necessita ser determinado em crianças com SNI para evitar efeitos adversos enquanto promove eficiência / We report a prospective study in ten children (mean age at presentation, 3.0 ± 1.6 years) using Cyclosporin(CSA) - a microemulsion to treat Idiopathic Nephrotic Syndrome (INS), with normal renal function evaluated by creatinine clearance estimated by stature, who presented C0 (trough level) between 50 and 150 ng/ml and achieved complete remission with CSA. The objective is to compare the pharmacokinetic parameters of CSA between nephrotic children on remission and relapse of the nephrotic state. The pharmacokinetic profile of CSA was evaluated with the 12-hour area under the time-concentration curve (AUC012) using seven time sample points. This procedure was done in the same patient during remission and relapse with the same dose of CSA (mg/kg). The AUC012 was calculated by trapezoidal rule. All points of the curve and the resumed AUC04 were compared with AUC012. We detected no significant differences at any points of the curve during remission and relapse, even when data were normalized by dose (mg/kg). Our study has shown that AUC04 was the main point on both states of the disease, remission and relapse, when compared to AUC012 (r=0.95 on remission and r=0.93 on relapse), and the second parameter found was C2 (r=0.86 on remission and r=0.80 on relapse). More controlled studies are needed to reproduce the findings and confirm this proposition. The target C2 concentration needs to be determined for INS in children to avoid adverse effects while promoting efficacy

Foi realizado um estudo prospectivo em dez crianças (média de idade de apresentação da doença, 3.0 ± 1.6 anos) usando ciclosporina (CSA) uma microemulsão utilizada no tratamento da síndrome nefrótica idiopática (SNI) - com função renal normal avaliada pelo clearance de creatinina estimado pela estatura, que apresentavam o C0 (nível do vale) entre 50 e 150 ng/ml e que encontraram remissão completa da doença com a CSA. O objetivo foi comparar os parâmetros farmacocinéticos da CSA entre crianças com SNI em remissão ou descompensadas da doença. O perfil farmacocinético da CSA foi avaliado através da área sob a curva de 12 horas (AUC012) utilizando sete pontos. Esse procedimento foi realizado no mesmo paciente durante a remissão e a descompensação da doença com a mesma dose de CSA em mg/kg. A AUC012 foi calculada através de régua trapezoidal. Todos os pontos da curva e a AUC04 simplificada foram comparados com a AUC012. Não foi encontrada diferença significativa em quaisquer pontos da curva nos dois estados da doença, nem mesmo quando normalizados pela dose em mg/kg. Nosso estudo mostrou que a AUC04 foi o principal parâmetro em ambos os estados da doença (remissão e recidiva) quando comparado com a AUC012 (r=0.95 na remissão e r=0.93 na recidiva), e o segundo melhor parâmetro foi o C2 (r=0.86 na remissão e r=0.80 na recidiva). Mais estudos controlados são necessários para reproduzir melhor estes achados e confirmar nossa proposta. Além disso, o C2 necessita ser determinado em crianças com SNI para evitar efeitos adversos enquanto promove eficiência / We report a prospective study in ten children (mean age at presentation, 3.0 ± 1.6 years) using Cyclosporin(CSA) - a microemulsion to treat Idiopathic Nephrotic Syndrome (INS), with normal renal function evaluated by creatinine clearance estimated by stature, who presented C0 (trough level) between 50 and 150 ng/ml and achieved complete remission with CSA. The objective is to compare the pharmacokinetic parameters of CSA between nephrotic children on remission and relapse of the nephrotic state. The pharmacokinetic profile of CSA was evaluated with the 12-hour area under the time-concentration curve (AUC012) using seven time sample points. This procedure was done in the same patient during remission and relapse with the same dose of CSA (mg/kg). The AUC012 was calculated by trapezoidal rule. All points of the curve and the resumed AUC04 were compared with AUC012. We detected no significant differences at any points of the curve during remission and relapse, even when data were normalized by dose (mg/kg). Our study has shown that AUC04 was the main point on both states of the disease, remission and relapse, when compared to AUC012 (r=0.95 on remission and r=0.93 on relapse), and the second parameter found was C2 (r=0.86 on remission and r=0.80 on relapse). More controlled studies are needed to reproduce the findings and confirm this proposition. The target C2 concentration needs to be determined for INS in children to avoid adverse effects while promoting efficacy

Para avaliar a exposição ao ácido micofenólico (MPA) na fase inicial pós-transplante renal, receptores foram destinados para receber tacrolimo (n=33) ou ciclosporina (n=19, controle) com MMF. Foram feitas coletas de farmacocinética (AUC) do inibidor de calcineurina e MPA nos dias 7, 14, 30, 60 e 180 pós-transplante. Dos dias 14-180, a MPA-AUC foi mais elevada no grupo tacrolimo devido a um maior segundo pico de MPA. Com doses fixas de MMF, uma grande porcentagem de curvas ficou abaixo da faixa terapêutica. No dia 7, a equação que emprega a concentração pré-dose (C0) e na segunda hora (C2) foi a mais precisa para estimar AUC. Após o dia 7, a equação que utiliza C2 foi a mais precisa. A exposição ao MPA nos primeiros seis meses após transplante renal é maior sob tacrolimo do que ciclosporina. Entretanto, para qualquer inibidor de calcineurina empregado com MMF, uma equação que emprega C0 e C2 (dia 7) e C2 isoladamente (após o dia 7), permite a monitoração de MPA com grande precisão / To evaluate the exposure to mycophenolic acid (MPA) early after renal transplantation, recipients were allocated to tacrolimus (n = 33) or Neoral (n =19, control) plus MMF. Pharmacokinetic curves (AUC) of calcineurin inhibitor and MPA were drawn on days 7, 14, 30, 60 and 180 post-transplant. From days 14-180, MPA-AUC was higher in tacrolimus group due to a higher second MPA peak. With fixed MMF doses, a great amount of curves fell below the proposed therapeutic range. On day 7, the equation that uses pre-dose (C0) and second-hour (C2) concentrations was the most accurate. After day 7, the equation that uses C2 alone was the most accurate. Exposure to MPA during the first six months after transplantation is higher under tacrolimus than Neoral. Nevertheless, despite the calcineurin inhibitor associated with MMF, an equation that uses C0 and C2 up to day 7 and C2 thereafter allows precise MPA monitoring

Equi-caloric bouts of interval (INT: 5x 2:2 min at 85 and 40% VO2max) and continuous (21 minutes at 60% VO2max) exercise were performed by 13 older prehypertensive males on separate days, at equivalent times of day, to assess the influence of exercise mode on post-exercise hypotension (PEH). Cardiovascular measures were collected for 30 min pre and 60 min post-exercise. PEH as measured by mean post-exercise systolic blood pressure (SBP) decrease, area under the SBP curve, and minimum SBP achieved, was equivalent after both conditions. SV was significantly reduced and HR was significantly elevated post-exercise after both conditions. No significant reductions in CO or TPR were observed. INT exercise provided a larger perturbation to the autonomic nervous system as measured by Baroreflex sensitivity and Heart Rate Variability. The responses elicited by acute INT exercise, with repeated exposure, may lead to greater improvements in blood pressure regulation than those associated with continuous aerobic training.

Equi-caloric bouts of interval (INT: 5x 2:2 min at 85 and 40% VO2max) and continuous (21 minutes at 60% VO2max) exercise were performed by 13 older prehypertensive males on separate days, at equivalent times of day, to assess the influence of exercise mode on post-exercise hypotension (PEH). Cardiovascular measures were collected for 30 min pre and 60 min post-exercise. PEH as measured by mean post-exercise systolic blood pressure (SBP) decrease, area under the SBP curve, and minimum SBP achieved, was equivalent after both conditions. SV was significantly reduced and HR was significantly elevated post-exercise after both conditions. No significant reductions in CO or TPR were observed. INT exercise provided a larger perturbation to the autonomic nervous system as measured by Baroreflex sensitivity and Heart Rate Variability. The responses elicited by acute INT exercise, with repeated exposure, may lead to greater improvements in blood pressure regulation than those associated with continuous aerobic training.

The receiver operating characteristic (ROC) curves is a popular tool for evaluating continuous diagnostic tests. The traditional definition of ROC curves incorporates implicitly the idea of "hard" thresholding, which also results in the empirical curves being step functions. The first topic is to introduce a novel definition of soft ROC curves, which incorporates the idea of "soft" thresholding. The softness of a soft ROC curve is controlled by a regularization parameter that can be selected suitably by a cross-validation procedure. A byproduct of the soft ROC curves is that the corresponding empirical curves are smooth. The second topic is on combination of several diagnostic tests to achieve better diagnostic accuracy. We consider the optimal linear combination that maximizes the area under the receiver operating characteristic curve (AUC); the estimates of the combination's coefficients can be obtained via a non-parametric procedure. However, for estimating the AUC associated with the estimated coefficients, the apparent estimation by re-substitution is too optimistic. To adjust for the upward bias, several methods are proposed. Among them the cross-validation approach is especially advocated, and an approximated cross-validation is developed to reduce the computational cost. Furthermore, these proposed methods can be applied for variable selection to select important diagnostic tests. However, the above best-subset variable selection method is not practical when the number of diagnostic tests is large. The third topic is to further develop a LASSO-type procedure for variable selection. To solve the non-convex maximization problem in the proposed procedure, an efficient algorithm is developed based on soft ROC curves, difference convex programming, and coordinate descent algorithm.

Para avaliar a exposição ao ácido micofenólico (MPA) na fase inicial pós-transplante renal, receptores foram destinados para receber tacrolimo (n=33) ou ciclosporina (n=19, controle) com MMF. Foram feitas coletas de farmacocinética (AUC) do inibidor de calcineurina e MPA nos dias 7, 14, 30, 60 e 180 pós-transplante. Dos dias 14-180, a MPA-AUC foi mais elevada no grupo tacrolimo devido a um maior segundo pico de MPA. Com doses fixas de MMF, uma grande porcentagem de curvas ficou abaixo da faixa terapêutica. No dia 7, a equação que emprega a concentração pré-dose (C0) e na segunda hora (C2) foi a mais precisa para estimar AUC. Após o dia 7, a equação que utiliza C2 foi a mais precisa. A exposição ao MPA nos primeiros seis meses após transplante renal é maior sob tacrolimo do que ciclosporina. Entretanto, para qualquer inibidor de calcineurina empregado com MMF, uma equação que emprega C0 e C2 (dia 7) e C2 isoladamente (após o dia 7), permite a monitoração de MPA com grande precisão / To evaluate the exposure to mycophenolic acid (MPA) early after renal transplantation, recipients were allocated to tacrolimus (n = 33) or Neoral (n =19, control) plus MMF. Pharmacokinetic curves (AUC) of calcineurin inhibitor and MPA were drawn on days 7, 14, 30, 60 and 180 post-transplant. From days 14-180, MPA-AUC was higher in tacrolimus group due to a higher second MPA peak. With fixed MMF doses, a great amount of curves fell below the proposed therapeutic range. On day 7, the equation that uses pre-dose (C0) and second-hour (C2) concentrations was the most accurate. After day 7, the equation that uses C2 alone was the most accurate. Exposure to MPA during the first six months after transplantation is higher under tacrolimus than Neoral. Nevertheless, despite the calcineurin inhibitor associated with MMF, an equation that uses C0 and C2 up to day 7 and C2 thereafter allows precise MPA monitoring