Στην παρούσα διπλωματική εργασία, πραγματοποιείται μία εκτεταμένη βιβλιογραφική αναφορά των πιο γνωστών και ευρέως χρησιμοποιούμενων μεθόδων για τον έλεγχο του βάθους της αναισθησίας (DoA) σε ασθενείς. Αν και ο χειροκίνητος έλεγχος της αναισθησίας παραμένει ο πιο διαδεδομένος τρόπος κατά τη διάρκεια των χειρουργικών επεμβάσεων, έχει πραγματοποιηθεί μία πληθώρα ερευνών και δοκιμών προκειμένου να αυτοματοποιηθεί η συγκεκριμένη διαδικασία. Στις περισσότερες μάλιστα από αυτές χρησιμοποιούνται μοντέλα PID ελεγκτών. Προκειμένου να επιτευχθούν καλύτερα επίπεδα εμπιστοσύνης και ακρίβειας, έχουν προταθεί διαφορετικοί τρόποι αντιμετώπισης που εφαρμόζονται πλέον σε ρεαλιστικά μοντέλα ασθενών. Επομένως, στη συγκεκριμένη εργασία γίνεται αναφορά στα διάφορα υπολογιστικά συστήματα που χρησιμοποιούνται επί του παρόντος για την υλοποίηση αλγορίθμων προβλεπτικού ελέγχου, περιγράφεται με λεπτομέρειες ο τρόπος εξαγωγής των Φαρμακοκινητικών και Φαρμακοδυναμικών μοντέλων, καθώς και επίσης μελετάται ο τρόπος λειτουργίας και η απόδοση διάφορων ελεγκτών, όπως απλός PID, MPC, RTDA και ενός deadbeat ελεγκτή, με σκοπό την εξομάλυνση της ύπνωσης έχοντας ως κύρια μεταβλητή ελέγχου την προποφόλη. / In this diploma thesis, there is an extended literature review of the most known and used methods for the depth of anesthesia (DoA) control systems. Although manual control of anesthesia is still the dominant practice during surgery, an increasing number of studies have been conducted to investigate the possibility of automating this procedure. These studies used proportional−integral−derivative (PID) controllers, as well as model-based controllers. However, there is a need for a comprehensive evaluation of closed-loop systems, to establish their safety, reliability, and efficacy for anesthesia regulation. This requires a detailed evaluation of promising and/or recent controllers for a range of patients and conditions via simulation. The present study investigates the different computational systems that are used to implement the appropriate algorithms for MPC ,the mathematic model of Pharmacokinetics and Pharmacodynamics models and the performance of single-loop PID, MPC, RTDA (Robustness, set point Tracking, Disturbance rejection, Aggressiveness) and deadbeat controllers for closed-loop regulation of hypnosis using propofol with bi spectral index (BIS) as the controlled variable

The purpose of this study is the comparison of methods of image enhancement kinetics in breast MRI tomography, according to 4 models, that analyze dynamic image series. Specifically, the following models have been implemented: (a) the Kuhl empirical approach, (b) a 3-parameter empirical model, (c) the 3 parameter mathematical model of Jansen and (d) the 5-parameter mathematical model of Fan. These models have been tested in a classification task of breast lesions (benignity/malignancy), using a k-ΝΝ (k=3 and k=7) classifier. A case sample of 29 benign and 49 malignant lesions, originating from 1.5T system, were analyzed. A graphical user interface has been implemented, intended as a visual aid to guide the identification of the location of the analyzing Region of Interest (ROI) of the lesion. In this study, the enhancement kinetic features of the two empirical models, as well as the primary and the secondary kinetic features of the two mathematical models were calculated. For proper ROI selection, 2 feature maps (a) the initial enhancement and (b) the 3 Time Point (3TP) kinetic map (Hauth et al. 2006), were utilized as pre-processing step. To evaluate the classification performance, indices such as sensitivity, specificity and accuracy were utilized. Employing the initial enhancement map, classification performance obtained for Kuhl empirical approach (Kuhl et al. 1999), the 3-parameter empirical model, the mathematical model of Jansen et al (Jansen et al. 2008) and the mathematical model of Fan (Fan et al. 2004, 2007) was: (0.87, 0.34, 67.9%), (0.81, 0.65, 70.5%), (0.85, 0.55, 70.5%) and (0.81, 0.58, 67.9%), respectively. Classification results employing the 3TP kinetic map for the Kuhl empirical approach (Kuhl 1999), 3-parameter empirical model, the mathematical model of Jansen and the mathematical model of Fan were: (0.95, 0.58, 82.0%), (0.95, 0.82, 84.6%), (0.85, 0.68, 78.2%) and (0.93, 0.79, 79.4%), respectively. In conclusion, the 3TP kinetic contributed in the proper location of the analyzing ROI and subsequently in the improved classification of malignant from benign lesions for all enhancement kinetic models studied. / Σκοπός της παρούσας εργασίας είναι η σύγκριση μεταξύ μεθόδων ανάλυσης της κινητικής του σκιαγραφικού στην μαγνητική τομογραφία μαστού, σύμφωνα με 4 μοντέλα ανάλυσης που αξιοποιούν τα απεικονιστικά δεδομένα δυναμικών ακολουθιών εικόνων. Συγκεκριμένα, υλοποιήθηκαν: (α) η εμπειρική προσέγγιση Kuhl et al. (Kuhl et al. 1999), η οποία χρησιμοποιεί 1 ποσοτικό δείκτη της αρχικής ενίσχυσης του σήματος και ποιοτική εκτίμηση της κινητικής του σκιαγραφικού στη φάση έκπλυσης, (β) ένα εμπειρικό μοντέλο 3 ποσοτικών δεικτών που ποσοτικοποιούν την πρόσληψη και έκπλυση, (γ) το 3-παραμετρικό εμπειρικό μαθητικό φαρμακοκινητικό μοντέλο των Jansen et al. (Jansen 2008), (δ) το 5-παραμετρικό εμπειρικό φαρμακοκινητικό μαθητικό μοντέλο των Fan et al. (Fan et al. 2004, 2007), στα πλαίσια ταξινόμησης αλλοιώσεων του μαστού (καλοήθεια/κακοήθεια) με χρήση ταξινομητή k-ΝΝ (k=3 και k=7). Μελετήθηκαν 29 καλοήθεις και 49 κακοήθεις αλλοιώσεις, και οι λήψεις των εικόνων έγιναν από 1.5 T μαγνητικό τομογράφο. Υλοποιήθηκε γραφικό περιβάλλον διεπαφής (Graphical User Interface-GUI), προτεινόμενο ως εργαλείο υποβοήθησης για την επιλογή της τοποθέτησης της περιοχής ενδιαφέροντος για την αξιολόγηση των κινητικών χαρακτηριστικών της αλλοίωσης. Στα πλαίσια της παρούσας μελέτης, υπολογίστηκαν τα κινητικά χαρακτηριστικά για τα δύο εμπειρικά μοντέλα καθώς και τα πρωτεύοντα και δευτερεύοντα χαρακτηριστικά για τα μαθηματικά μοντέλα. Για την επιλογή της περιοχής ενδιαφέροντος υλοποιήθηκαν: (α) ένας κινητικός χάρτης πρώιμης ενίσχυσης σήματος, (β) ο κινητικός χάρτης 3TP (Hauth et al. 2006), οποίος εκφράζει τη συνολική κινητική του σκιαγραφικού. Για την αξιολόγηση της απόδοσης ταξινόμησης (διαφοροποίηση καλοήθειας/κακοήθειας) χρησιμοποιήθηκαν οι δείκτες ευαισθησία, ειδικότητα και ακρίβεια. Με χρήση του χάρτη πρώιμης ενίσχυσης για την επιλογή της περιοχής ενδιαφέροντος, η απόδοση ταξινόμησης, του εμπειρικού μοντέλου Kuhl et al. (1999), του εμπειρικού μοντέλου των 3 παραμέτρων, του μαθηματικού μοντέλου Jansen και του μαθηματικού μοντέλο Fan (Fan et al. 2004, 2007) ήταν: (0.87, 0.34, 67.9%), (0.81, 0.65, 70.5%), (0.85, 0.55, 70.5%) και (0.81, 0.58, 67.9%), αντιστοίχως. Με χρήση του κινητικού χάρτη 3TP η απόδοση ταξινόμησης του εμπειρικού μοντέλου Kuhl (Kuhl et al. 1999), του εμπειρικού μοντέλου των 3 παραμέτρων, του μαθηματικού φαρμακοκινητικού μοντέλου Jansen et al. (Jansen et al. 2008) και του μαθηματικού φαρμακοκινητικού μοντέλου Fan (Fan et al. 2004, 2007) ήταν: (0.95, 0.58, 82.0%), (0.95, 0.82, 84.6%), (0.85, 0.68, 78.2%) και (0.93, 0.79, 79.4%), αντιστοίχως. Συμπερασματικά, χρήση του κινητικού χάρτη 3TP συνεισφέρει σε ορθότερη επιλογή της θέσης της περιοχής ενδιαφέροντος προς ανάλυση, βελτιώνοντας αποτελέσματα της ταξινόμησης των κακοηθών από καλοήθεις αλλοιώσεις για όλα τα μοντέλα κινητικής σκιαγραφικού που μελετήθηκαν.

Orientador: José Ricardo de Arruda Miranda / Resumo: As nanopartículas magnéticas (NPMs) são uma classe de nanopartículas que se destacam em áreas da saúde, principalmente em aplicações teranósticas. O potencial das NPMs é prejudicado em função da absorção hepática, considerando que o fígado é uma rede complexa de células inter-relacionadas responsável pela captação do NPMs. Apesar de vários estudos concentrados na área, ainda é pouco compreendido como cada estrutura hepática opera no processo de retirada das NPMs da circulação sanguínea. As técnicas de imagem tem proporcionado avanços no entendimento de eventos fisiológicos, facilitando sua visualização. Atualmente, há uma gama de modalidades responsáveis pela detecção e imagiamento da biodistribuição das NPMs. Dentre essas técnicas de imagens, várias estão presentes apenas em hospitais ou em grandes centros de pesquisa. Esse trabalho tem como destaque, a aplicação de um novo método biomagnético, a Biosusceptometria AC (BAC), para a detecção in vivo de NPMs através de imagens. Fatores como ausência de radiação ionizante, versatilidade e alta resolução temporal são atributos do sistema frente às diversas técnicas de imagem. Neste estudo foi utilizado um novo arranjo do sistema, o sistema Multicanal BAC (ACB-MC), o qual permitiu a avaliação do processo de clearence das NPMs no sangue e seu posterior acúmulo no fígado em ratos Wistar. A partir das imagens dinâmicas obtidas e sua quantificação, foi proposto uma abordagem matemática a fim de auxiliar a farmacocinética de distribuiç... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: Magnetic nanoparticles (MNPs) are widely used in healhty research areas, especially in terantic applications. The potential use of MNPS is diminished due to hepatic absorption, considering that the liver is a complex network of interrelated cells and the most responsible by MNPs uptake. Although there has been many studies in this area, it is still unknown how the liver removes MNPs from the bloodstream. Imaging techniques have provided more understanding of physiological events, enhancing their visualization. Currently, there is a range of modalities able to detect and image the biodistribution of MNPs. Among these techniques, several are restricted to large hospitals and research centers. In this work, our aim was to highlight the application of a new biomagnetic method, Biosusceptometry AC (BAC), for in vivo detection of MNPs through images. When compared to other systems, BAC has some advantages such as the absence of ionizing radiation, versatility and high temporal resolution. In this study, a new system arrangement, the Multicanal BAC system (ACB-MC) was used. This allowed the evaluation of the clearence process of MNPs in the blood and their subsequent accumulation in the liver. Those experiments were performed in Wistar rats. From the quantification of the dynamic images, a mathematical approach was proposed to investigate the pharmacokinetics of distribution and accumulation of MNPs. The ACB-MC system presented excellent temporal resolution and through sequential im... (Complete abstract click electronic access below) / Mestre

Resistant bacteria are becoming more and more of a problem but treatment with antibiotics in combination may overcome and prevent resistance development. The combination of meropenem and colistin against Pseudomonas aeruginosa has been proposed as a promising treatment to increase the bactericidal effect and a synergistic effect has been proposed. A Pharmacokinetic-Parmacodynamic (PKPD) model that describes the dynamics of bacteria kill could be used to evaluate if the effects are additive or not. The model could later also be used to find optimal dosing for both of the antibiotics used alone or in combination with each other. The aim of the present study was to develop a PKPD model that describes the bactericidal activity of the two antibiotics, both in mono-therapy and in combination. The data were from in vitro static time kill-curve experiments that had been conducted on two strains of Pseudomonas aeruginosa; the wild-type (ATCC 27853) and the resistant-type (PL0603761). Resistance was observed in the experimental data and thus it had to be taken into account in the modelling. PKPD models were fitted to the bacterial counts in NONMEM with pharmacodynamic compartments for susceptible and resting bacteria. In the resting compartment the bacteria could not be killed. The bacteria moved into the resting compartment from the susceptible compartment when a certain concentration of bacteria was obtained. A pharmacokinetic compartment characterized changes in drug concentrations and the drug degradation during the experimental time was considered. Two different drug effects were tried on the susceptible bacteria, linear effect and Emax models.. The resistance development occurring during the experiments was described by two compartments where the parameter kon determined the rate of onset of resistance development. In the final model, kon was found to either be concentration-independent or dependent, depending on antibiotics and bacteria. The degree of resistance development produced an overall inhibitory effect on the drug effect. The growth rate was estimated to be lower and the EC50 to be higher for the resistant compared to the wild-type bacteria. The model was used to predict the expected time-kill curve if the effect of the two drugs are additive when combining the two drugs. The observed  bacteria kill was lower than the model predicted for the wild-type bacteria. For the resistant bacteria the assumption of additive bacteria kill for the two drugs-seemed adequate.

Tato dizertační práce představuje metodu zobrazování perfúze magnetickou rezonancí, jež je výkonným nástrojem v diagnostice, především v onkologii. Po ukončení sběru časové sekvence T1-váhovaných obrazů zaznamenávajících distribuci kontrastní látky v těle začíná fáze zpracování dat, která je předmětem této dizertace. Je zde představen teoretický základ fyziologických modelů a modelů akvizice pomocí magnetické rezonance a celý řetězec potřebný k vytvoření obrazů odhadu parametrů perfúze a mikrocirkulace v tkáni. Tato dizertační práce je souborem uveřejněných prací autora přispívajícím k rozvoji metodologie perfúzního zobrazování a zmíněného potřebného teoretického rozboru.

Pharmacokinetic and pharmacodynamic models can be applied to clinical study data using various modelling approaches depending on the aim of the analysis. In population pharmacokinetics for instance, simple compartmental models can be employed to describe concentration-time data, identify prognostic factors and interpolate within well-defined experimental conditions. The first objective of this thesis was to illustrate such a ‘semi-mechanistic’ pharmacokinetic modelling approach using mavoglurant as an example of a compound under clinical development. In particular, methods to accurately characterise complex oral pharmacokinetic profiles and evaluate the impact of absorption factors were investigated. When the purpose of the model-based analysis is to further extrapolate beyond the experimental conditions in order to guide the design of subsequent clinical trials, physiologically-based pharmacokinetic (PBPK) models are more valuable as they incorporate information not only on the drug but also on the system, i.e. on mammillary anatomy and physiology. The combination of such mechanistic models with statistical modelling techniques in order to analysis clinical data has been widely applied in toxicokinetics but has only recently received increasing interest in pharmacokinetics. This is probably because, due to the higher complexity of PBPK models compared to conventional pharmacokinetic models, additional efforts are required for adequate population data analysis. Hence, the second objective of this thesis was to explore methods to allow the application of PBPK models to clinical study data, such as the Bayesian approach or model order reduction techniques, and propose a general mechanistic modelling workflow for population data analysis. In pharmacodynamics, mechanistic modelling of clinical data is even less common than in pharmacokinetics. This is probably because our understanding of the interaction between therapeutic drugs and biological processes is limited and also because the types of data to analyse are often more complex than pharmacokinetic data. In oncology for instance, the most widely used clinical endpoint to evaluate the benefit of an experimental treatment is survival of patients. Survival data are typically censored due to logistic constraints associated with patient follow-up. Hence, the analysis of survival data requires specific statistical techniques. Longitudinal tumour size data have been increasingly used to assess treatment response for solid tumours. In particular, the survival prognostic value of measures derived from such data has been recently evaluated for various types of cancer although not for pancreatic cancer. The last objective of this thesis was therefore to investigate different modelling approaches to analyse survival data of pancreatic cancer patients treated with gemcitabine, and compare tumour burden measures with other patient clinical characteristics and established risk factors, in terms of predictive value for survival.

Acute liver failure (ALF) is a rare but devastating clinical syndrome with multiple causes and a variable course. The mortality rate is high. Orthotopic liver transplantation is the only therapy of proven survival benefit but the limited supply of donor organs, the rapidity of progression and the variable course of ALF limit its use. A need therefore exists for a method to ‘bridge’ patients, that is, provide temporary support, to either the spontaneous regeneration of the innate liver or transplantation. One possibility includes bio-artificial liver support systems (BALSS). This technology is composed of an extracorporeal circulation system incorporating a bioreactor that contains parenchymal liver cells (hepatocytes) to perform the detoxifying, transforming and synthetic properties of a liver. However, the development of a BALSS holds particular challenges. Despite approximately four decades of research, bio-artificial liver (BAL) technology globally remains in a pre-commercial stage. The University of Pretoria (UP) and the Council for Scientific and Industrial Research (CSIR) have developed a BALSS with novel characteristics. These include a computationally optimized radial-flow primary porcine hepatocyte bioreactor perfused with blood plasma, and a perfluorocarbon oxygen carrier which replaces hemoglobin. There are also novel design properties in the circulation system itself. Demonstrating the metabolic and clinical efficacy of a BAL device requires implementing, in vitro (cell biology), in vivo (animal) and mathematical modeling studies. These studies are a formal necessity but are inherently ‘models’ of the in vivo human clinical circumstance. That is, they are limited by their experimentally controlled configuration/s. In investigating these, this thesis firstly provides a foundation by reviewing the clinical and biological context of ALF and BAL technology, then presents and evaluates particular studies/models that have been implemented over several years in the course of the UP-CSIR BAL project. For each section, thoughts and recommendations regarding future work that will facilitate the development of BAL technology are discussed in detail. The thesis is concluded with an evaluation of success and the consensus-agreed requirement of continued research and innovation in the field. / Thesis (PhD)--University of Pretoria, 2010. / Chemical Engineering / unrestricted

This thesis deals with perfusion analysis using DCE-MRI (Dynamic contrast-enhanced magnetic resonance imaging). DCE-MRI is commonly used for microcirculation evaluation mainly in oncology and in recent years also in cardiology. The theoretical overview focuses on the issue of pharmacokinetic modeling and the estimation of perfusion parameters using selected models. The experimental part describes research software PerfLab and then it is aimed at the proposed program for synthetic data generation. Simulated data obtained under ideal conditions and in the presence of noise were used to compare models for the accuracy and reliability of DCE-MRI analysis.

This thesis deals with the principles of ultrasonic perfusion analysis and methods for determining perfusion parameters. It examines Evolutionary algorithms and their ability to optimize the approximation of dilution curves from ultrasond tissue scannig. It compares the optimization performance of three evolutionary algorithms. Continuous genetic algorithm GA, algorithm SOMA and PSO. Methods are evaluated on simulated and clinical data.