Fluidized bed technology is used for thermal conversion of solid fuels (combustion and gasification) and is especially suitable for conversion of low-rank fuels such as biomass and waste. The performance of fluidized bed units depends on the fuel mixing and fuel-gas contact. Thus, it is important to understand these two phenomena in order to develop models for reliable design and scale up of fluidized bed units. This work investigates, under conditions representative for industrial fluidized bed units, the lateral fuel mixing (in a unit with a cross section of 1.44 m2 both at hot and cold conditions) and the bubble flow distribution (in a 1.2 m-wide 2-dimensional unit). The work confirms previous findings on the formation of preferred bubble paths and shows that these bubble paths are enhanced by lowering the fluidization velocity, increasing the dense bed height and reducing the pressure drop across the gas distributor. From the fuel mixing experiments, an estimation of the lateral effective dispersion coefficient to values in the order of 10-3 m2/s is obtained under both hot and cold conditions. The experiments under cold conditions give additional qualitative information on the fuel mixing patterns such as flotsam/jetsam tendencies. The camera probe developed for fuel tracking under hot conditions enables to study the fuel dispersion under real operation at relevant industrial scales. Based on the characteristics of the bubble path flow, a model for the horizontal fuel dispersion on a macroscopic scale is formulated and shown to be able to give a good description of the experimental data. As opposed to the commonly applied diffusion-type modeling of the lateral solids dispersion, the proposed model facilitates integration with models of the bubble flow. Thus, the present modeling work is a first step to provide a modeling of the fuel dispersion, which uses as inputs only the main operational parameters of the fluidized bed.

No description available.

Automated tissue image analysis aims to develop algorithms for a variety of histological applications. This has important implications in the diagnostic grading of cancer such as in breast and prostate tissue, as well as in the quantification of prognostic and predictive biomarkers that may help assess the risk of recurrence and the responsiveness of tumors to endocrine therapy. In this thesis, we use pattern recognition and image analysis techniques to solve several problems relating to histopathology and immunohistochemistry applications. In particular, we present a new method for the detection and localization of tissue microarray cores in an automated manner and compare it against conventional approaches. We also present an unsupervised method for color decomposition based on modeling the image formation process while taking into account acquisition noise. The method is unsupervised and is able to overcome the limitation of specifying absorption spectra for the stains that require separation. This is done by estimating reference colors through fitting a Gaussian mixture model trained using expectation-maximization. Another important factor in histopathology is the choice of stain, though it often goes unnoticed. Stain color combinations determine the extent of overlap between chromaticity clusters in color space, and this intrinsic overlap sets a main limitation on the performance of classification methods, regardless of their nature or complexity. In this thesis, we present a framework for optimizing the selection of histological stains in a manner that is aligned with the final objective of automation, rather than visual analysis. Immunohistochemistry can facilitate the quantification of biomarkers such as estrogen, progesterone, and the human epidermal growth factor 2 receptors, in addition to Ki-67 proteins that are associated with cell growth and proliferation. As an application, we propose a method for the identification of paired antibodies based on correlating probability maps of immunostaining patterns across adjacent tissue sections. Finally, we present a new feature descriptor for characterizing glandular structure and tissue architecture, which form an important component of Gleason and tubule-based Elston grading. The method is based on defining shape-preserving, neighborhood annuli around lumen regions and gathering quantitative and spatial data concerning the various tissue-types.

Poverty and long distances are two reasons why some people in the third world countries hasdifficulties seeking medical help. A solution to the long distances could be if the medical carewas more mobile and diagnostically tests could be performed on site in villages. A new pointof-care test based on a small blood shows promising results both in run time and mobility.However, the method still needs more advanced equipment for analysis of the resultingmicroarray. This study has investigated the potential to perform the analysis within asmartphone application, performing all steps from image capturing to a diagnostic result. Theproject was approach in two steps, starting with implementation and selection of imageanalysis methods and finishing with implementing those results into an Android application.A final application was not developed, but the results gained from this project indicates that asmartphone processing power is enough to perform heavy image analysis within a sufficientamount of time. It also imply that the resolution in the evaluated images taken with a Nexus 6together with an external macro lens most likely is enough for the whole analysis, but furtherwork must be done to ensure it.

Thesis (MScEng (Process Engineering))--University of Stellenbosch, 2010. / Thesis presented in partial fulfilment of the requirements for the degree MASTER OF SCIENCE IN ENGINEERING (EXTRACTIVE METALLURGICAL ENGINEERING) in the Department of Process Engineering at the University of Stellenbosch / ENGLISH ABSTRACT: Flotation is an important processing step in the mineral processing industry wherein valuable minerals are extracted. Flotation is a difficult process to control due to its complexity, meaning that the reversal of series of changes will not necessarily bring the process back to its original state. Expert knowledge is incorporated in flotation control through operator experience and intervention, which is subject to many challenges, creating the need for improvement in control. The performance of a flotation cell is often determined by evaluating froth appearance. The application of image analysis to capture, evaluate and monitor froth appearance poses multiple benefits such as consistent and reliable froth appearance evaluation. The objective for this study was to conduct a laboratory study for the collection of froth images with the purpose of evaluating the feasibility of using image information to predict platinum froth grade. Laboratory test work was performed according to a fractional factorial experimental design. Six variables were considered: air flowrate, pulp level and collector, activator, frother and depressant dosages. The laboratory study results were quantified by assay analysis. Analysis of variance only revealed the significant effect of pulp height and collector addition on flotation performance. Data pre-processing revealed information regarding feature correlations and variance contributions. Data analysis from captured images achieved reliable froth grade predictions using random forest classification and artificial neural network (ANN) regression techniques. Random forest classification accuracies of 86.8% and 75.5% were achieved for the following respective datasets: image data of each individual experiment (average of all experiments) and all image data. The applied ANN models achieved R2 values 0.943 and 0.828 for the same 2 datasets. An industrial case study was done wherein a series of step changes in air flowrate was made on a specific flotation cell. The limited industrial case study results supported laboratory study results. Multiple linear regression performed very well, reaching Rª values up to 0.964. Neural networks achieved slightly better with R2 values of up to 0.997. Based on the findings, the following main conclusions were drawn from this study: - Reliable predictions using classification and regression models on image data were proved possible in concept by the laboratory study, and supported by results from an industrial case study on a narrow system. The following main recommendations were made for further investigation: - Research over a larger range of operating conditions is needed to find a more comprehensive solution. - Investigations should be conducted to determine hardware requirements and specifications in terms of minimum resolution, lighting requirements, sampling frequency and data storage. Software requirements, specifications and maintenance challenges should also be investigated for implementation purposes once a more comprehensive solution has been found. - Strategies in terms of camera placement and model building will need to follow, giving special attention to a strategy to handle ore composition change. / AFRIKAANSE OPSOMMING: Flotasie is ‘n belangrike proses in die mineraal proseseringsbedryf vermoeid met die ontginning van waardevolle minerale. Die proses is moelik om te beheer vanweë sy kompleksiteit, wat verwys na die onvermoë om die proses terug te bring na sy oorspronklike toestand deur ‘n reeks veranderinge om te keer. In die algemeen word spesialis kennis deel van prosesbeheer deur die toepassing van operateurs se ervaring en ingryping, wat opsigself verskeie uitdagings bied wat die behoefte aan verbeterde beheertoestelle en strategieë daarstel. Die werkverrigting van flotasieselle word gereeld beoordeel op grond van die voorkoms van die skuim. Die gebruik van beeldverwerking om dié inligting vas te vang vir monitering en evaluering doeleindes hou verskeie voordele in, bv. konsikwente en betroubare evaluasie van die skuimvoorkoms. Die doelwitte vir hierdie studie was om ‘n laboratorium studie te loods vir die opname van skuimbeelde, met die doel om die bruikbaarheid van beeldinligting vir die voorspelling van die flotasieprodukkwaliteit, te ondersoek. Die laboratorium gevallestudie is uitgevoer aan die hand van ‘n fraksionele faktoriale eksperimentele ontwerp. Ses veranderlikes was ondersoek naamlik, lugvloeitempo, pulphoogte en versamelaar aktiveerder en depressant toevoeging. Die studie se resultate is gekwantifiseer deur die analise van die skuim inhoud. ‘n Analise van variansie het slegs die invloed van pulphoogte en versamelaartoevoeging op die flotasievertoning uitgelig. Data voorverwerking het inligting uitgelig rondom die veranderlikes se verhouding met mekaar. Data analise metodes, naamlik lukrake klassifiseringswoude en neurale netwerk regressie, is toegepas op die versamelde beelddata en het belowende resultate gelewer. Lukrake klassifiseringswoude het klasse gedentifiseer met akkuraathede van 86.8% en 75.5% vir die volgende onderskeie datastelle: individuele eksperimente se beeld data (gemiddeld oor alle eksperimentele lopies), alle beelddata as een stel. Die neurale netwerke het Rª waardes van 0.943 rn 0.828 gelewer vir dieselfde 2 datastelle. Die beperkte nywerheidsgevallestudie het verandering in lugvloeitempo toegelaat vir ‘n enkele flotasie sel. Die resultate het die bevindinge van die laboratorium gevallestudie gesteun. Veelvoudige lineere regressie het Rª waardes van tot en met 0.964 gelewer. Neurale netwerke het daarop verbeter met waardes tot en met 0.997. Die volgende hoof gevolgtrekkinge was duidelik vanuit die resultate: - Betroubare voorspellings was moontlik met die toepassing van klassifikasie en regressie modelle op die laboratorium studie data. Die resultate is ondersteun deur soortgelyke resultate van die beperkte nywerheidsgevallestudie. Die volgende hoof aanbevelings was gemaak vir verdere navorsing: - Navorsing oor ‘n wyer reeks proseskondisies is nodig om ‘n meer omvattende oplossing te vind. - ‘n Ondersoek moet geloods word om die hardeware vereistes en spesifikasies in terme van die minimum beeld resolusie, beligting vereistes, monsterneming tempo en die berging van data te bepaal. Sagteware vereistes, spesifikasies en instandhouding uitdagings moet ook ondersoek word vir implementasie doeleindes sodra ‘n meer omvattende oplossing gevind is. - Strategieë in verband met die plasing van kamers en die ontwikkeling van modelle is nodig, waarin spesiale aandag gegee moet word om die probleem van veranderende ertssamestelling op te los.

Background: Brain metastasis is a leading cause of cancer mortality and affects 20-40% of all cancer patients. The BBB is responsible for isolation and protection of the brain parenchyma from many diagnostic and therapeutic agents. New molecular agents that target tumour-associated VCAM-1 expression on the brain endothelium show improvements in the early diagnosis of brain metastasis. The vascular endothelium of the CNS plays an important role in the maintenance of the brain microenvironment and possibly aids the extravasation of tumour cells via expression of CAMs. Aims: Using the breast carcinoma-derived 4T1 cell line, syngeneic to BALB/c mice, this work aimed (i) to determine the level of colocalisation between VCAM-1 expression at sites of brain metastasis and the presence of VCAM-MPIO-induced hypointensities in MR datasets; (ii) to describe the normal developmental characteristics of the intracardial BALB/c-4T1 brain metastatic model in the absence of overt inflammation; (iii) to test the effects of an adenovirus-induced systemic inflammatory challenge on metastatic uptake and development in the brain. Results: The level of correspondence of VCAM-MPIO-derived hypointensities with VCAM-1 expression at the tumour site was found to be dependent on the size of metastasis. An improved method for detection of VCAM-MPIO hypointensities using an automated method has been presented. Tumours were found to develop preferentially on venous rather than arteriolar blood vessels, and showed greater and lesser abundance in different anatomical brain regions. Adenovirus injection was found to cause an upregulation of a range of peripheral pro-inflammatory cytokines, and expression of VCAM-1 on cerebral vasculature, preferentially on arteriolar blood vessels. Both pre- and post-treatment with adenovirus caused a two-fold reduction in tumour numbers and altered developmental characteristics of established tumours, although no significant differences were observed in VCAM-MPIO hypointensities in MR datasets. Conclusions: The development of molecular MRI approaches to target VCAM-1 expression at the site of brain metastases has improved the sensitivity of tumour detection. 4T1-GFP metastasis to the brain is specific both to anatomical sites and to regions of the vascular bed, suggesting differences in vascular morphology and/or signalling dynamics in these locations. The changes in tumour number and morphology as a result of systemic inflammation suggest an anti-tumour effect of adenoviral treatment and, given the role of the systemic immune system and its importance in the development of immunotherapies, possible future directions for research.

The aim of this research was to address the computational complexity in designing multimodality Computer-Aided Diagnosis (CAD) systems for characterising breast lesions, by harnessing the general purpose computational potential of consumer-level Graphics Processing Units (GPUs) through parallel programming methods. The complexity in designing such systems lies on the increased dimensionality of the problem, due to the multiple imaging modalities involved, on the inherent complexity of optimal design methods for securing high precision, and on assessing the performance of the design prior to deployment in a clinical environment, employing unbiased system evaluation methods. For the purposes of this research, a Pattern Recognition (PR)-system was designed to provide highest possible precision by programming in parallel the multiprocessors of the NVIDIA’s GPU-cards, GeForce 8800GT or 580GTX, and using the CUDA programming framework and C++. The PR-system was built around the Probabilistic Neural Network classifier and its performance was evaluated by a re-substitution method, for estimating the system’s highest accuracy, and by the external cross validation method, for assessing the PR-system’s unbiased accuracy to new, “unseen” by the system, data. Data comprised images of patients with histologically verified (benign or malignant) breast lesions, who underwent both ultrasound (US) and digital mammography (DM). Lesions were outlined on the images by an experienced radiologist, and textural features were calculated. Regarding breast lesion classification, the accuracies for discriminating malignant from benign lesions were, 85.5% using US-features alone, 82.3% employing DM-features alone, and 93.5% combining US and DM features. Mean accuracy to new “unseen” data for the combined US and DM features was 81%. Those classification accuracies were about 10% higher than accuracies achieved on a single CPU, using sequential programming methods, and 150-fold faster. In addition, benign lesions were found smoother, more homogeneous, and containing larger structures. Additionally, the PR-system design was adapted for tackling other medical problems, as a proof of its generalisation. These included classification of rare brain tumours, (achieving 78.6% for overall accuracy (OA) and 73.8% for estimated generalisation accuracy (GA), and accelerating system design 267 times), discrimination of patients with micro-ischemic and multiple sclerosis lesions (90.2% OA and 80% GA with 32-fold design acceleration), classification of normal and pathological knee cartilages (93.2% OA and 89% GA with 257-fold design acceleration), and separation of low from high grade laryngeal cancer cases (93.2% OA and 89% GA, with 130-fold design acceleration). The proposed PR-system improves breast-lesion discrimination accuracy, it may be redesigned on site when new verified data are incorporated in its depository, and it may serve as a second opinion tool in a clinical environment.

This thesis addresses three research gaps in published literature. These are, the absence of Object Based Image Analysis (OBIA) methods for urban Land Use and Land Cover (LULC) analysis in Nigeria; the inability to use Nigeriasat-1 satellite data for urban LULC analysis and monitoring urban growth in Nigeria with Shannon’s Entropy Index. Using Abuja as a case study, this research investigated the nature of land use/land cover change (LULCC). Specific objectives were: design of an object based classification method to extract urban LULC; validate a method to extract LULC in developing countries from multiple sources of remotely sensed data; apply the method to extract LULC data; use the outputs to validate an Urban Growth Model (UGM); optimise an UGM to represent patterns and trends and through this iterative process identify and prioritise the driving forces of urban change; and finally use the outputs of the land use maps to determine if planning has controlled land use development. Landsat 7 ETM (2001), Nigeriasat-1 SLIM (2003) and SPOT 5 HRG (2006) sensor data were merged with land use cadastre in OBIA, to produce land use maps. Overall classification accuracies were 92%, 89% and 96% respectively. Post classification analysis of LULCC indicated 4.43% annual urban spread. Shannon’s Entropy index for the study period were 0.804 (2001), 0.898 (2003) and 0.930 (2006). Cellular Automata/Markov analysis was also used to predict urban growth trend of 0.89% per annum. For the first time OBIA has been used for LULC analysis in Nigeria. This research has established that Nigeriasat-1 data can contribute to urban studies using innovative OBIA methods. In addition, that Shannon’s Entropy Index can be used to understand the nature of urban growth in Nigeria. Finally, the drivers of LULCC in Abuja are similar to those of planned capital cities in other developing economies. Land use developments in Abuja can provide an insight into urban dynamics in a developing country’s capital region. OBIA, Shannon’s Entropy Index and UGM can aid urban administrators and provide information for sustainable urban planning and development.

This thesis is about channel-coded feature maps applied in view-based object recognition, tracking, and machine learning. A channel-coded feature map is a soft histogram of joint spatial pixel positions and image feature values. Typical useful features include local orientation and color. Using these features, each channel measures the co-occurrence of a certain orientation and color at a certain position in an image or image patch. Channel-coded feature maps can be seen as a generalization of the SIFT descriptor with the options of including more features and replacing the linear interpolation between bins by a more general basis function. The general idea of channel coding originates from a model of how information might be represented in the human brain. For example, different neurons tend to be sensitive to different orientations of local structures in the visual input. The sensitivity profiles tend to be smooth such that one neuron is maximally activated by a certain orientation, with a gradually decaying activity as the input is rotated. This thesis extends previous work on using channel-coding ideas within computer vision and machine learning. By differentiating the channel-coded feature maps with respect to transformations of the underlying image, a method for image registration and tracking is constructed. By using piecewise polynomial basis functions, the channel coding can be computed more efficiently, and a general encoding method for N-dimensional feature spaces is presented. Furthermore, I argue for using channel-coded feature maps in view-based pose estimation, where a continuous pose parameter is estimated from a query image given a number of training views with known pose. The optimization of position, rotation and scale of the object in the image plane is then included in the optimization problem, leading to a simultaneous tracking and pose estimation algorithm. Apart from objects and poses, the thesis examines the use of channel coding in connection with Bayesian networks. The goal here is to avoid the hard discretizations usually required when Markov random fields are used on intrinsically continuous signals like depth for stereo vision or color values in image restoration. Channel coding has previously been used to design machine learning algorithms that are robust to outliers, ambiguities, and discontinuities in the training data. This is obtained by finding a linear mapping between channel-coded input and output values. This thesis extends this method with an incremental version and identifies and analyzes a key feature of the method -- that it is able to handle a learning situation where the correspondence structure between the input and output space is not completely known. In contrast to a traditional supervised learning setting, the training examples are groups of unordered input-output points, where the correspondence structure within each group is unknown. This behavior is studied theoretically and the effect of outliers and convergence properties are analyzed. All presented methods have been evaluated experimentally. The work has been conducted within the cognitive systems research project COSPAL funded by EC FP6, and much of the contents has been put to use in the final COSPAL demonstrator system.

Advances in genetics and fluorescent protein chemistry have enabled us to fuse fluorescent probes directly to biomolecules in stably growing organisms; making it easier to image the precise position and movement of cells in three dimensions. Fluorescent stains and dyes can be employed in a similar fashion to visualize nano-scale fluctuations in active cellular structures without fixation. While informative and exciting on a qualitatively level, microscopy truly becomes powerful when we can extract meaningful quantitative information. To accomplish this, custom MATLAB (Mathworks, Natick, MA) image analysis algorithms were developed to specifically measure the biophysical parameters related to pathogenesis and function in microbes and mammalian cells. These parameters can then be exploited in the development of biophysical models to validate current measurements, and make critical predictions about the system's behavior, often addressing quantities inaccessible by experimental methods. The following research chapters of this dissertation thoroughly describe how these techniques were developed and applied to study the biophysical mechanisms of bacterial and cellular invasion.