No description available.

In 2012, more than 1.6 million new cases of breast cancer were diagnosed and about half a million women died of breast cancer. The incidence has increased in the developing world. The mortality, however, has decreased. This is thought to partly be the result of advances in diagnosis and treatment. Studying tissue samples from biopsies through a microscope is an important part of diagnosing breast cancer. Recent techniques include camera-equipped microscopes and whole slide scanning systems that allow for digital high-throughput scanning of tissue samples. The introduction of digital pathology has simplified parts of the analysis, but manual interpretation of tissue slides is still labor intensive and costly, and involves the risk for human errors and inconsistency. Digital image analysis has been proposed as an alternative approach that can assist the pathologist in making an accurate diagnosis by providing additional automatic, fast and reproducible analyses. This thesis addresses the automation of conventional analyses of tissue, stained for biomarkers specific for the diagnosis of breast cancer, with the purpose of complementing the role of the pathologist. In order to quantify biomarker expression, extraction and classification of sub-cellular structures are needed. This thesis presents a method that allows for robust and fast segmentation of cell nuclei meeting the need for methods that are accurate despite large biological variations and variations in staining. The method is inspired by sparse coding and is based on dictionaries of local image patches. It is implemented in a tool for quantifying biomarker expression of various sub-cellular structures in whole slide images. Also presented are two methods for classifying the sub-cellular localization of staining patterns, in an attempt to automate the validation of antibody specificity, an important task within the process of antibody generation.  In addition, this thesis explores methods for evaluation of multimodal data. Algorithms for registering consecutive tissue sections stained for different biomarkers are evaluated, both in terms of registration accuracy and deformation of local structures. A novel region-growing segmentation method for multimodal data is also presented. In conclusion, this thesis presents computerized image analysis methods and tools of potential value for digital pathology applications.

Many disorders are associated with regional brain volumes. The analysis of these volumes from MR images often requires sequential processing steps such as localization and delineation. It is common to perform volumetric normalization using intracranial volume (ICV, the total volume inside the cranial cavity) when comparing regional brain volumes, since head size varies considerably between individuals. Multiple methods for estimating ICV and procedures for volume normalization exist. A method for interhemispheric surface localization and extraction, using both intensity and symmetry information and without time consuming pre-processing, was developed. Evaluations of hemisphere division accuracy as well as suitability as a pre-processing step for interhemispheric structure localization were made. The performance of the method was comparable to that of methods focusing on either of these tasks, making it suited for use in many different studies. Automated ICV estimations from Freesurfer and SPM were evaluated using 399 reference segmentations. Both methods overestimated ICV and estimations using Freesurfer contained errors associated with skull-size. Estimations from SPM contained errors associated with gender and atrophy. An experiment showed that the choice of method can affect study results. Manual ICV estimation is very time consuming, but can be performed using only a subset of voxels in an image to increase speed and decrease manual labor. Segmenting every nth slice and stereology were evaluated in terms of required manual labor and estimation error, using the previously created ICV references. An illustration showing how much manual labor is required for a given estimation error using different combinations of n and stereology grid spacing was presented. Finally, different procedures for ICV normalization of regional brain volumes when investigating gender related volume differences were theoretically explained and evaluated using both simulated and real data. Resulting volume differences were seen to depend on the procedure used. A suggested workflow for procedure selection was presented. Methodological contributions that can aid the analysis of the human brain have been presented. The performed studies also contribute to the understanding of important methodological considerations for regional brain volume analysis.

Since its introduction in 1960's, the model organism Caenorhabditis elegans has played a crucial role towards scientific discoveries because of its relatively simple anatomy, conserved biological mechanisms, and mapped genome. The organism also has a rapid generation time and produces a large number of isogenic progeny, making C. elegans an excellent system for conducting forward genetic screens. Conventional screening methods, however, are labor intensive and introduce potential experimental bias; typically, large-scale screens can take months to years. Thus, automated screening and characterization platforms can provide an opportunity to overcome this bottleneck. The objective of this thesis is to develop tools to perform rapid phenotypical characterization of C. elegans to enable automated genetic screening systems for neural development. To achieve this goal, I developed methods to increase throughput of worm handling using microfluidic devices and demonstrate software modules to phenotype unknown mutants using quantitative and morphological image analysis methods. Microfluidic devices are constructed from PDMS using established soft-lithography techniques. The emphasis on the simplification of existing designs greatly facilitates the adoption of our developed systems by other scientists. This thesis also includes image processing modules using various techniques to determine animal phenotypes. For example, we adapted standard thresholding methods to detect animal motor neurons, developed a modified granulometry algorithm to rapidly characterize large numbers of lipid droplets in 3D, and developed a probability model to determine neuronal process morphology. This work is significant because it increases current capabilities of screening small animals with morphological phenotypes by enhancing throughput and reducing human bias. Genes or gene functions that can be discovered using these methods can further elucidate mechanisms relevant to neural development, degeneration, maintenance, and function; these discoveries in turn can facilitate discoveries of potential therapeutic strategies for human neurological diseases.

The object of this thesis is to present two novel applications of Spatiotemporal Image Correlation Spectroscopy (STICS) to biological systems. STICS is a technique which uses the correlations in pixel intensity fluctuations of an image time series, captured under fluorescence microscopy, to measure the speed and direction of a flowing population of fluorescently labeled molecules. The method was first applied to measure the dynamics of transport vesicles inside growing pollen tubes of lily flowers. The measured vector maps allowed to confirm the presence of actin filaments along the periphery of the tubes, as well as the presence of a reverse-fountain pattern in the apical region. In a second set of experiments, STICS was used to measure the retrograde flow of filamentous actin in migrating chick DRG neuronal growth cones. These results serve as proof of principle that STICS can be used to probe the response of the growth cone cytoskeleton to external chemical cues.

New Zealand�s main export earnings come from the primary production area including agriculture, horticulture, and viticulture. One of the major contributors in this area of horticulture is the production of quality export grade fruit; specifically apples. In order to maintain a competitive advantage, the systems and methods used to grow the fruit are constantly being refined and are increasingly based on data collected and analysed by both the orchardist who grows the produce and also researchers who refine the methods used to determine high levels of fruit quality. To support the task of data analysis and the resulting decision-making process it requires efficient and reliable tools. This thesis attempts to address this issue by applying the techniques of Connectionist-Based Intelligent Information Systems (CBIIS) for Image Analysis and Knowledge Discovery. Using advanced neurocomputing techniques and a novel knowledge engineering methodology, this thesis attempts to seek some solutions to a set of specific problems that exist within the horticultural domain. In particular it describes a methodology based on previous research into neuro-fuzzy systems for knowledge acquisition, manipulation, and extraction and furthers this area by introducing a novel and innovative knowledge-based architecture for knowledge-discovery using an on-line/real-time incremental learning system based on the Evolving Connectionist System (ECOS) paradigm known as the Evolving Fuzzy Neural Network (EFuNN). The emphases of this work highlights knowledge discovery from these data sets using a novel rule insertion and rule extraction method. The advantage of this method is that it can operate on data sets of limited sizes. This method can be used to validate the results produced by the EFuNN and also allow for greater insight into what aspects of the collected data contribute to the development of high quality produce.

Thesis (Ph.D.) -- McMaster University, 2003. / Includes bibliographical references (leaves 134-141). Also available via World Wide Web.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniv., 2004. / Härtill 4 uppsatser.

Diss. Uppsala : Sveriges lantbruksuniv., 2004.

Diss. (sammanfattning) Uppsala : Sveriges lantbruksuniversitet, 2004. / Härtill 6 uppsatser.