FE - Teses de Doutoramento / Doctoral Theses
Permanent URI for this collection
Browse
Recent Submissions
- Multilayer perceptron modelling of industrial cultivation processes for DNA vaccine production control and optimizationPublication . Silva, Daniel Alexandre Felício da; Silva, Tito Lívio dos Santos; Calado, Cecília Ribeiro da CruzBioinformatics is one of the emerging fields in Computer Sciences with an increasingly more impactful role, in a symbiotic-like association with Biology and Bioprocesses, aiding in the knowledge of complex mechanisms and elements in both these fields of knowledge. The permanent and exponential development of equipments that allow large-scale data acquisition, has set-in-motion the need to create methodologies to store and analyze that same data, in a manner that researchers can extract its meaning, with a high degree of confidence and precision, in a viable time-frame, and therefore help them in their research. However, there are techniques and procedures where the possibility of extracting detailed information in real-time is limited, either by the absence of adequate equipment or by the logistics’ impossibility often times associated with a thorough gathering of information regarding those processes. The industrial cultivation process is one of those cases where environmental values such as Dissolved Oxygen, pH, Temperature and others are available in real-time but the information regarding the complex molecular constituents of the cultivation are missing, only being obtainable by off-line analysis. In industry, as a way of minimizing the risk of contaminations, the number of samples collected for analysis along the fermentative process is always kept at a minimal level and can even be non-existent in most cases. Because of this, the real knowledge on the cultivation process is limited, most of the times, to the initial state of the cultivation and its final state, as obtaining exact readings along the cultivation is quite difficult. All control decisions on the system are based on indirect evaluations as the rate of oxygen consumption or the pH variation. This limited knowledge may impair the reproducibility of the cultivation process, as cells are living organisms that present a natural variability. That natural variability is further enhanced by slight variations of the environmental cultivation conditions. This is crucial in case of biopharmaceutics production due to the high regulatory constraints. DNA plasmid vaccines are increasingly moving to the forefront of pharmaceutical products due to their potential advantage over viral vectors, and due to the theoretical advantages of DNA-vaccines over subunit and whole cells vaccines. The plasmid vaccine production consists in the growth, in bioreactors, of bacteria such as Escherichia coli containing the plasmid vector with engineered DNA that is afterwards extracted. However, as previous referred, it is highly relevant to control the whole cultivation process, as there is still a great need for process optimization. This optimization can result in high-yield production with reduced production costs. One solution that has been presented for this kind of control and optimizations is based on computational simulation of the processes. Computer simulations, or in silico, are often used to quickly test multiple scenarios without the need to allocate specific resources, human or material, as for instance high-cost reagents, turning these processes into more viable ones. With the more complex work occurring in the early stages of model development. Furthermore, mathematical models may also be useful to estimate, along extensive periods of time, the complex molecular constituents of a cultivation process by using the real time analysis gathered by the sensors generally used in industry. This work’s objective is to use computational methodologies to determine the behavior of a recombinant E. coli culture designed to produce plasmids for DNA vaccination. This work was performed in the Engineering Faculty of the Catholic University of Portugal and the Instituto de Medicina Molecular. In this work we propose the use of a Multilayer Perceptron (MLP) in order to monitor and understand the behavior of E.coli DH5- containing the vector pVAX-LacZ plasmid, during different batch and fed-batch cultivations. The focus of this work consisted in studying the behavior of cultivations with different initial pre-set conditions concerning the carbon-source, pH and feeding strategy, and with intermediate perturbations determined experimentally. With this goal, a set of cultivations were defined as examples, in order to allow us to explore a wide universe in terms of variables, as well as establishing a comparison between cultivations with similar initial conditions. MLPs are part of a larger universe referred as Artificial Neural Networks (ANNs). They are considered universal approximators, allowing for the identification of complex patterns by learning training examples. These examples will influence future state predictions. This characteristic allows a great adaptability to different models as the main limitation of MLPs is centered in the quality and quantity of training examples, rather than pre-determined functions and parameters. Unlike conventional modelling techniques, MLPs rely on the data rather than theoretical assumptions. This means that the possibility of introducing bias in the pattern recognition is less likely. Moreover, MLPs can serve as hypothesis validators. In this work we were able to obtain model fit values (R2) that in most cases were superior to 0.7. These values are even more interesting when we take into account the number of variables we attempted to predict and cross it with the number of training examples we were able to produce. In order to achieve our goal we defined the following real time and off-line variables. The off-line variables were: concentration of Biomass, Plasmid, Glucose, Glycerol and Acetate. These variables are not quantified in real time, as it is required to extract a sample from the bioreactor and subsequently analyzed it. The on-line variables, acquired in real time were: Dissolved Oxygen Concentration, pH, Stirring Rate and Feeding Rate. The technical and logistics inability to quantify each variable at the same exact rate illustrates two fundamental issues with the basic cultivation monitoring process: the standardization of the moment in which the variables are quantified; and the determination of the next state of the cultivation. In this work, we establish that the prediction was made using 1-hour spaced intervals using a cross-validation training methodology. This 1-hour spacing was determined by analyzing data available from Martins (2008) and observing no significant increase in network prediction with 15 minute, 30 minute or 60 minute intervals. Finally, we present a possible methodology for optimizing fed-batch cultivations based on Genetic Algorithms (GA). In this approach, information and parameters of the trained MLP are used to create a cultivation policy that will be applied during the industrial process. Genetic Algorithms are evolutionary algorithms based on computational adaptations of biological evolutionary theories. Our Genetic Algorithm approach is based on a chromosome representation of a decision tree designed to determine the course of experimental action according to the state of the controlled variables. These evaluations are based on the values of Glycerol, Glucose and Acetate and according to their values a feeding rate is determined for the next time-point. This methodology in an early stage could allow the definition of a wider example space and then translate into a cultivation strategy closer to the optimal solution. This research work aims to answer this emerging need and contribute to the advance of the knowledge in the area, opening new paths for further research that natural and desirably will follow.
- Hybrid-Trefftz finite elements for elastostatic and elastodynamic problems in porous mediaPublication . Toan, Cao Duc; Moldovan, Ionut Dragos; Freitas, João António Teixeira deThe displacement and stress models of the hybrid-Trefftz finite element formulation are applied to the elastostatic and elastodynamic analysis of two-dimensional saturated and unsaturated porous media problems. The formulation develops from the classical separation of variables in time and space, but it leads to two time integration strategies. The first is applied to periodic problems, which are discretized in time using Fourier analysis. A mixed finite element approach is used in the second strategy for discretization in time of non-periodic/transient problems. These strategies lead to a series of uncoupled problems in the space dimension, which is subsequently discretized using either the displacement or the stress model of the hybrid-Trefftz finite element formulation. The main distinction between the two models is in the way that the interelement continuity is enforced. The displacement model enforces the interelement compatibility, while the stress model enforces the interelement equilibrium. As is typical of Trefftz methods, for both models, the approximation bases are constrained to satisfy locally the homogeneous form of the domain (Navier) equations. The free-field solutions of these equations are derived in cylindrical coordinates and used to construct the domain approximations of the hybrid-Trefftz displacement and stress elements. If the original equations are non-homogeneous, the influence of the source terms is modelled using Trefftz-compliant solutions of the corresponding static problem. For saturated porous media, the finite element models are based on the Biot's theory. It assumes an elastic solid phase fully permeated by a compressible liquid phase obeying the Darcy's law. For the modelling of unsaturated porous media, the finite elements are formulated using the theory of mixtures with interfaces. The model is thermodynamically consistent and considers the full coupling between the solid, fluid and gas phases, including the effects of relative (seepage) accelerations. Small displacements and linear-elastic material behaviour are assumed for all models.
- Digital mammography : characterisation of practice and equipment performance in portuguese healthcare providersPublication . Reis, Cláudia Isabel Sá dos; Pascoal, Ana Isabel Lourenço; Alves, João Henrique GarciaMammography is a key imaging modality used in screening and front line diagnosis of breast disease. Digital imaging systems for mammography were introduced in the 1980´s and since then they have been replacing the conventional analogue (screen-film) systems. The adequate clinical performance of digital mammography has been validated in extensive trials and the modality gained popularity and is currently in clinical use worldwide. The successful transition from screen-film to digital mammography and the cost-effective use of the digital technology requires a clear understanding of the potentials and limitations of digital imaging systems and the impacts on established workload and workflow routines. The adoption of optimised practices in digital mammography is a critical step to produce images with adequate diagnostic quality whilst keeping the radiation dose (and associated risk to the patient) to the minimum necessary. Digital mammography was introduced in Portugal in 2000 and currently is in use in various hospitals and clinics for screening, diagnostic and intervention. However, the existing literature is limited and obsolete and does not allow a clear, and independent, perspective of the level of innovation in mammography in Portugal nor the quality of mammography services provided to the patient. This research investigates: the level of innovation in mammography in Portugal, the practices in use with focus on the technology and technical performance of digital mammography systems installed in Portuguese healthcare providers. In addition, compliance with international quality standards for mammography is assessed and, where appropriate, recommendations are provided to improve current practice. An extensive search was performed in governmental bodies, healthcare providers and industry to map the installed base of mammography technology in Portugal. A survey to healthcare providers was undertaken to capture data on the practices in use for mammography in hospitals and clinics. A comprehensive technical protocol was developed for assessing the performance of digital mammography systems. The existing guidance European Reference Organisation for Quality Assured Breast Screening and Diagnostic Services (EUREF), International Atomic Energy Agency (IAEA), Institute of Physics and Engineering in Medicine (IPEM) and National Healthcare Service Breast Screening Program (NHSBSP) were considered. This protocol was applied to a sample (52) of digital mammography systems in current use in hospitals and clinics in Portugal and included tests to the performance of the X-ray source and to the image receptor. A total of 446 mammography systems were identified in Portugal (including conventional and digital technology) with the vast majority (76%) installed in private healthcare centers. The equipment age in clinical use varied noticeably ranging from legacy to state of the art equipment. The majority of mammography systems (64%) are computed radiography (CR) imaging technology based on a photostimuable phosphor image receptor (image plate) and an associated optical processor. The number of direct digital mammography systems (DR) is relevant (67) and has been gradually increasing. Two state of the art mammography systems with tomosynthesis capabilities were identified. Sixty five healthcare providers participated in the survey to map practices in digital mammography. The survey showed that the selection of exposure settings in digital mammography equipment is most frequently driven by the recommendations of the manufacturers. A limited number of centers (26%) stated using international guidelines as reference. The use of protocols developed locally was rarely reported. A routine quality control (QC) programme is in place in 70% of the participants’ centers and it is undertaken most frequently under an outsourcing contract by external QC providers. The support of a local medical physicist/engineer for QC and optimization activities is rare in the participant centers. The vast majority (76%) of radiographers participating in the survey does not perform periodic analysis of rejected (and repeated) mammography images, despite this being a simple and important QC task. Radiographers and radiologists involved in mammography reported training needs in digital mammography. Radiographers would welcome training in artifact recognition, quality control and dosimetry. Radiologists named training needs also in artifact recognition and quality control and, additionally, in breast pathology. The majority (97%) of radiologists agrees with the need to implement a program for certification of mammography in Portugal and would adhere to a voluntary program. Arguments presented were that certification could bring necessary improvements in staff training whilst contributing to maintain equipment performance and staff practices at adequate levels. In addition certification could promote quality and harmonization of mammography practice nationwide. The technical assessment of the mammography systems showed a very wide range of performance across the various systems. The performance of the X-ray system (tube and generator) was generally very good for most systems. The tube output was accurate and reproducible across the range of clinically relevant tube voltage (kVp) and tube charge (mAs) settings. A very large proportion of mammography systems tested (70%) showed misalignment between the light and radiation fields. For 8 systems the misalignment is causing unnecessary irradiation beyond the breast anatomy towards the patient chest wall and lungs. The majority of image receptors (83% CR and 93% DR) showed good uniformity. Artifacts were observed on various CR image plates some of which pose the risk to compromise the visibility of clinically relevant details. These were due to dust in the phosphor, damaged phosphor due to extensive use and aging, and mechanical stress during processing. Measurements of Signal-difference to Noise Ratio (SdNR) for typical clinical settings (AEC) were performed for various breast thicknesses (mimicked with slabs of Perspex). For DR systems the vast majority of results obtained were within the recommended reference values (IAEA) with exceptions observed for the larger breast thicknesses. For CR systems the mean values of SdNR were frequently lower than the reference values suggesting a possible miscalibration of the AEC system or limited contrast performance of the image receptor. SdNR should be assessed regularly as it is an important objective metric of the performance of the image receptor and has an impact in image contrast. The mean glandular dose (MGD) was measured on a standard Perspex phantom (45 mm) for each system using the method proposed by Dance et al. In addition MGD was measured in realistic (clinical) conditions using exposure settings collected from mammography examinations. Dose reference levels (DRL) were determined using the 75th percentile of the MGD distributions. For both measurements with Perspex and using clinical exposures, the average MGD for CR systems was higher compared to the DR systems. The average MGD in phantom (considering both technologies) is within the acceptable reference range proposed by EUREF (< 2.5 mGy). The mean MGD for the mediolateral oblique (MLO) projection was higher than for the craniocaudal (CC) projection. For CR systems the results are 1.85 mGy (CC projection) and 2.10 mGy (MLO projection). For DR systems the corresponding values are 1.54 mGy (CC) and 1.68 mGy (MLO). The DRL (clinical) is 2.2 mGy and 1.5 mGy for CR and DR systems, respectively. The DRL (phantom) is 2.0 mGy and 1.2 mGy for CR and DR systems, respectively. The technical image quality (IQ) performance was assessed with images produced with a breast phantom (TORMAM). Approximately 200 images were acquired and analysed visually by 4 trained observers on a calibrated high-resolution electronic display. The analysis consisted on the identification of low contrast detail detectability thresholds for structures in the phantom image that mimic breast abnormalities (spicules, nodules, clusters of micro-calcifications). An overall average score was calculated for each image and used as a metric of image quality. A wider range of IQ scores was observed for CR mammography systems (30-100) compared to DR systems (61-91) showing wider variations in image quality between systems. The 75th percentile of the score distribution was score=70 for CR and score=85 for DR revealing a lower IQ performance of CR systems compared to DR systems. A small but important number of mammography systems (8 CR and 4 DR) showed very poor technical IQ performance (percentile 25th) and require further investigation. For both technologies higher IQ scores were most frequently associated with higher MGD values with some exceptions observed. For comparative IQ scores the MGD is higher for CR systems compared to DR systems showing improved dose efficiency of the DR technology. The research has documented important data and conclusions regarding the installed base of mammography equipment in Portugal, current practices and technical performance of mammography systems in clinical use. The need of dedicated training in mammography was highlighted by radiologists and radiographers involved in mammography. The study provides a background to all stakeholders for planning improvements to mammography departments.