No description available.

No description available.

We introduce new methods for preprocessing and normalization of data acquired by gas/liquid chromatography coupled with mass spectrometry (GC/LC-MS). Normalization is desired prior to subsequent statistical analysis to adjust variabilities in ion intensities that are not caused by biological differences. There are different sources of experimental bias including variabilities in sample collection, sample storage, poor experimental design, noise, etc. Also, instrument variability in experiments involving a large number of runs leads to a significant drift in intensity measurements. We propose new normalization methods based on bootstrapping, Gaussian process regression, non-negative matrix factorization (NMF), and Bayesian hierarchical models. These methods model the bias by borrowing information across runs and features. Another novel aspect is utilizing scan-level data to improve the accuracy of quantification. We evaluated the performance of our method using simulated and experimental data. In comparison with several existing methods, the proposed methods yielded significant improvement. Gas chromatography coupled with mass spectrometry (GC-MS) is one of the technologies widely used for qualitative and quantitative analysis of small molecules. In particular, GC coupled to single quadrupole MS can be utilized for targeted analysis by selected ion monitoring (SIM). However, to our knowledge, there are no software tools specifically designed for analysis of GS-SIM-MS data. We introduce SIMAT, a new R package for quantitative analysis of the levels of targeted analytes. SIMAT provides guidance in choosing fragments for a list of targets. This is accomplished through an optimization algorithm that has the capability to select the most appropriate fragments from overlapping peaks based on a pre-specified library of background analytes. The tool also allows visualization of the total ion chromatogram (TIC) of runs and extracted ion chromatogram (EIC) of analytes of interest. Moreover, retention index (RI) calibration can be performed and raw GC-SIM-MS data can be imported in netCDF or NIST mass spectral library (MSL) formats. We evaluated the performance of SIMAT using several experimental data sets. Our results demonstrate that SIMAT performs better than AMDIS and MetaboliteDetector in terms of finding the correct targets in the acquired GC-SIM-MS data and estimating their relative levels. / Ph. D.

In this study, the impact of multiple image preprocessing methods on Convolutional Neural Networks (CNN) was studied. Metrics such as accuracy, precision, recall and F1-score (Hossin et al. 2011) were evaluated. Specifically, this study is geared towards malaria classification using the data set made available by the U.S. National Library of Medicine (Malaria Datasets n.d.). This data set contains images of thin blood smears, where uninfected and parasitized blood cells have been segmented. In the study, 3 CNN models were proposed for the parasitization classification task. Each model was trained on the original data set and 4 preprocessed data sets. The preprocessing methods used to create the 4 data sets were grayscale, normalization, histogram equalization and contrast limited adaptive histogram equalization (CLAHE). The impact of CLAHE preprocessing yielded a 1.46% (model 1) and 0.61% (model 2) improvement over the original data set, in terms of F1-score. One model (model 3) provided inconclusive results. The results show that CNN’s can be used for parasitization classification, but the impact of preprocessing is limited. / I denna studie studerades effekten av flera bildförbehandlingsmetoder på Convolutional Neural Networks (CNN). Mätvärden såsom accuracy, precision, recall och F1-score (Hossin et al. 2011) utvärderades. Specifikt är denna studie inriktad på malariaklassificering med hjälp av ett dataset som tillhandahålls av U.S. National Library of Medicine (Malaria Datasets n.d.). Detta dataset innehåller bilder av tunna blodutstryk, med segmenterade oinfekterade och parasiterade blodceller. I denna studie föreslogs 3 CNN-modeller för parasiteringsklassificeringen. Varje modell tränades på det ursprungliga datasetet och 4 förbehandlade dataset. De förbehandlingsmetoder som användes för att skapa de 4 dataseten var gråskala, normalisering, histogramutjämning och kontrastbegränsad adaptiv histogramutjämning (CLAHE). Effekten av CLAHE-förbehandlingen gav en förbättring av 1.46% (modell 1) och 0.61% (modell 2) jämfört med det ursprungliga datasetet, vad gäller F1-score. En modell (modell 3) gav inget resultat. Resultaten visar att CNN:er kan användas för parasiteringsklassificering, men effekten av förbehandling är begränsad.

Reliable results are crucial when working with medical decision support systems. A decision support system should be reliable but also be interpretable, i.e. able to show how it has inferred its conclusions. In this thesis, the preprocessing perceptron is presented as a simple but effective and efficient analysis method to consider when creating medical decision support systems. The preprocessing perceptron has the simplicity of a perceptron combined with a performance comparable to the multi-layer perceptron. The research in this thesis has been conducted within the fields of medical informatics and intelligent computing. The original idea of the production line as a tool for a domain expert to extract information, build decision support systems and integrate them in the existing system is described. In the introductory part of the thesis, an introduction to feed-forward neural networks and fuzzy logic is given as a background to work with the preprocessing perceptron. Input to a decision support system is crucial and it is described how to gather a data set, decide how many and what kind of inputs to use. Outliers, errors and missing data are covered as well as normalising of the input. Training is done in a backpropagation-like manner where the division of the data set into a training and a test set can be done in several different ways just as the training itself can have variations. Three major groups of methods to estimate the discriminance effect of the preprocessing perceptron are described and a discussion of the trade-off between complexity and approximation strength are included. Five papers are presented in this thesis. Case studies are shown where the preprocessing perceptron is compared to multi-layer perceptrons, statistical approaches and other mathematical models. The model is extended to a generalised preprocessing perceptron and the performance of this new model is compared to the traditional feed-forward neural networks. Results concerning the preprocessing layer and its connection to multivariate decision limits are included. The well-known ROC curve is described and introduced fully into the field of computer science as well as the improved curve, the QROC curve. Finally a tutorial to the program trainGPP is presented. It describes how to work with the preprocessing perceptron from the moment when a data file is provided to the moment when a new decision support system is built.

This thesis deals with problems of recognition of an electrometer´s state from sensing image. It is tangibly about electrometer´s scanning by a mobile phone´s camera. There is a surface with an electrometer´s dial which is detected and on this surface the particular numbers are detected consequently. The numbers are recognized via neural network. For more information from this image there are used some techniques of image segmentation to check the status. For the classification of the segmentation´s outputs are used classification tools, especially a support vector machine (SVM) and neural networks. Problems of image segmentations are solved by using OpenCV library. OpenCV is used for the implementation of the vector machine either. Application is on Android platform. Part of the thesis is concerned in a creation of a desktop application which is instrumental towards testing of neural network. The thesis also describes how to save the necessary data gathering in the course of the recognition which are used for working with neural network. The part of the thesis also deals with running web which will be evolved for the opportunity to participate in the further development of the system. There is available a public repository with source codes created during implementation.

Prediction of energy load profiles is an important topic in Smart Grid technologies. Accurate forecasts can lead to reduced costs and decreased dependency on commercial power suppliers by adapting to prices on energy market, efficient utilisation of solar and wind energy and sophisticated load scheduling. This thesis compares various statistical and machine learning models and their ability to forecast load profile for an entire day divided into 48 half-hour intervals. Additionally, we examine various preprocessing methods and their influence on the accuracy of the models. We also compare a variety of imputation methods that are designed to reconstruct missing observation commonly present in energy consumption data.

With arrival of affordable digital technology we are increasingly coming into contact with digital images. Cameras are no longer dedicated devices, but part of almost every mobile phone, PDA and laptop. This paper discusses methods for enhancing quality of digital images with focus on removing noise, creating high dynamic range (HDR) images and extending depth of field (DOF). It contains familiarization with technical means for acquiring digital image, explains origin of image noise. Further attention is drawn to HDR, from explaining the term, physical basis, difference between HDR sensing and HDR displaying, to survey and historical development of methods dealing with creating HDR images. The next part is explaining DOF when displaying, physical basis of this phenomenon and review of methods used for DOF extension. The paper mentions problem of acquiring images needed for solving given tasks and designs method for acquiring images. Using it a database of test images for each task was created. Part of the paper also deals with design of a program, that implements discussed methods, for solving the given tasks. With help of proposed class imgmap, quality of output images is improved, by modifying maps of input images. The paper describes methods, improvements, means of setting parameters and their effects on algorithms and control of program using proposed GUI. Finally, comparison with free software for extending DOF takes place. The proposed software provides at least comparable results, the correct setting of parameters for specific cases allows to achieve better properties of the resulting image. Time requirements of image processing are worse because designed software was not optimised.

Clustering of text data is one of tasks of text mining. It divides documents into the different categories that are based on their similarities. These categories help to easily search in the documents. This thesis describes the current methods that are used for the text document clustering. From these methods we chose Simultaneous keyword identification and clustering of text documents (SKWIC). It should achieve better results than the standard clustering algorithms such as k-means. There is designed and implemented an application for this algorithm. In the end, we compare SKWIC with a k-means algorithm.

In order to prepare modern finite element analysis a program for the efficient parallel generation of tetrahedral meshes in a wide class of three dimensional domains having a generalized cylindric shape is presented. The applied mesh generation strategy is based on the decomposition of some 2D-reference domain into single con- nected subdomains by means of its triangulations the tetrahedral layers are built up in parallel. Adaptive grid controlling as well as nodal renumbering algorithms are involved. In the paper several examples are incorporated to demonstrate both program's capabilities and the handling with.