Tesis Doctorales

Examinar

Mostrando 1 - 2 de 2

Development and validation of data analysis automation methods using pattern recognition
(Universidad Rey Juan Carlos, 2020) Pardo Sánchez, Esteban
Data analysis automation is an area of growing interest thanks to the increasing need of processing large amounts of data in a timely fashion, the large volumes of labeled data generated collectively, and the recent technological advances that enabled widespread adoption of multicore computing. This thesis explores three main areas where analysis automation has been proven essential. First, the field of cellular astronomy is studied. Cell astronomy is a specific type of low magnification imaging cytometry where fluorescent samples are imaged with a magnification such that cells have only a couple of pixels in radius. While significantly increasing the field of view and enabling cheap and quick analysis of thousands of cells, cell astronomy introduces some important challenges. These challenges include the detection of bright spots at low signal to noise ratios (SNRs), the estimation of cell diameter in the presence of partial volume effects, and the estimation of fluorescence intensity despite local background fluorescence. Fortunately, cell astronomy images resemble both astronomy and superresolution microscopy images, so popular image analysis methods in these fields can be used to overcome some of the main challenges. Using these fields as inspiration, a novel image analysis pipeline was created, which estimates both fluorescence intensity and cell diameter by fitting an heterogeneous mixture model using expectation maximization. This method is explained thoroughly in the included journal publication, which also validates the proposed pipeline using cell controls and microbeads. Second, we explore the task of automatic chromosome identification. Chromosomes can be imaged using a technique called multiplex fluorescence in situ hybridization, where chromosomes are labeled using at least five different fluorescent probes, and captured using multispectral imaging. Despite the developments in chromosome labeling, the analysis of this images still remains a manual or semaiautomated approach, where karyotyping is performed using both spectral and spatial information. Due to the recent popularity of convolutional networks, and the reach of near human performance for multiple tasks, we theorize that image segmentation using convolutional networks can achieve state of the art results for the analysis of multispectral chromosome images. To prove this, we have published a paper where a convolutional approach for chromosome identification is proposed. The attached journal publication describes an end to end segmentation network for the interpretation of multispectral chromosome images which uses both spectral and spatial information. The proposed method was evaluated using a publicly available dataset, outperforming previous automated methods, and achieving an average correct classification ratio (CCR) that has only been previously achieved using semiautomated approaches. Third, we investigate seismic phase picking automation. Phase picking deals with the identification of the arrival times of seismic waves, which is usually performed manually, or in a semiautomated fashion. The process of performing manual picks is described, underlining how this cumbersome process is often overlooked leading to intra-, and intersubject biases. Additionally, while there are widely available algorithms that automate this task, they are dated and do not offer the performance necessary to fully offload the job. On the other hand, while convolutional network approaches have been proposed for the analysis of seismic phases, we show some important issues that arise when directly applying regression or segmentation networks. In order to overcome this issues, we propose a two stage convolutional network where the first step computes a rough segmentation mask, a the second step computes a distance map to pinpoint the precise location, and then both steps are combined using an adaptation of the Hough transform. The proposed network was evaluated on publicly available data collected by the Northern California Earthquake Data Center (NCEDC), achieving a mean absolute error lower than previously proposed convolutional networks. Finally, an additional chapter includes some smaller contributions. On the one hand, an air quality forecasting method is presented. This method uses long short-term memory (LSTM) units to analyze a time series comprised by both air quality, and meteorological information. Then, the method is compared with Caliope, a model based air quality forecasting method, achieving lower mean squared error for the open data published by the city of Madrid. On the other hand, we study wood conductivity assessment using xylem cross sections. Traditionally, a specific set of hand tuned parameters would be necessary to analyze each tree species. This thesis shows that convolutional networks can learn the features used to segment conductive elements and ring paths of multiple tree species simultaneously. Additionally, a web application, Xyat (https://xyat.app), has been developed to enable researchers to use the proposed method without installing any software.
Measurement-Based Model Estimation for Deformable Objects
(Universidad Rey Juan Carlos, 2014-09) Miguel Villalba, Eder
Deformable objects play a critical role in our life due to their compliance. Clothing and support structures, such as mattresses, are just a few examples of their use. They are so common that an accurate prediction of their behavior under a variety of environments and situations is mandatory in order to design products with the desired functionalities. However, obtaining realistic simulations is a difficult task. Both, an appropriate deformation model and parameters that produce the desired behavior must be used. On one hand, there exist many deformation models for elasticity, but there are few capable of capturing other complex effects that are critical in order to obtain the desired realism. On the other hand, the task of estimating model parameters is usually performed using a trial-and-error method, with the corresponding waste in time. In this thesis we develop novel deformation models and parameter estimation methods that allow us to increase the realism of deformable object simulations. We present deformation models that capture several of these complex effects: hyperelasticity, extreme nonlinearities, heterogeneities and internal friction. In addition, we design parameter estimation methods that take advante of the structure of the measured data and avoid common problems that arise when numerial optimization algorithms are used. First, we focus on cloth and present a novel measurement system that captures the behavior of cloth under a variety of experiments. It produces a complete set of information including the 3D reconstruction of the cloth sample under test as well as the forces being applied. We design a parameter estimation pipeline and use this system to estimate parameters for several popular cloth models and evaluate their performance and suitability in terms of quality of the obtained estimations. We then develop a novel, general and flexible deformation model based on additive energy density terms. By using independent components this model allows us to isolate the effect that each one has on the global behavior of the deformable object, replicate existing deformation models and produce new ones. It also allows us to apply incremental approaches to parameter estimation. We demonstrate its advantages by applying it in a wide variety of scenarios, including cloth simulation, modeling of heterogeneous soft tissue and capture of extreme nonlinearities in finger skin. Finally, a fundamental observation extracted from the estimation of parameters for cloth models is that, in real-world, cloth hysteresis has a huge effect in the mechanical behavior and visual appearance of cloth. The source of hysteresis is the internal friction produced by the interactions between yarns. Mechanically, it can produce very different deformations in the loading or unloading cycles, while visually, it is responsible for effects such as persistent deformations, preferred wrinkles or history-dependent folds. We develop an internal friction model, present a measurement and estimation system that produces elasticity and internal friction parameters, and analyse the visual impact of internal friction in cloth simulation.

Examinar

Examinando Tesis Doctorales por Materia "1203.18 Sistemas de Información, Diseño Componentes"