|
|
3D reconstruction from multiple images
Řehoř, Jan ; Odstrčilík, Jan (referee) ; Jakubíček, Roman (advisor)
This thesis deals with 3D reconstruction of objects from multiple frames. Thesis is divided into five chapters. The first one is connected with theory, so, it is about 3D reconstruction, about camera geometry description and about reconstruction algorithms. The following are made up by practical parts, in which are described single steps of 3D reconstruction realization, so, it is about frame acquisition, camera calibration and about used algorithms. Last part of the thesis is made up by discussion and conclusion in which the results of the experiment are discussed and summarized.
|
|
|
Detection and segmentation of lumbar vertebrae in 3D CT data
Nemček, Jakub ; Kolář, Radim (referee) ; Jakubíček, Roman (advisor)
This thesis deals with the detection and the segmentation of lumbar vertebrae in CT image datas. The described detection method is based on the use of a trained SVM classificator and histograms of oriented gradients as the image features. The detection method is applied on two-dimensional sagital slices of the CT image. The segmentation method is implemented as triangular mesh model deformation of models, that are obtained from averaged vertebrae in real CT datas. The first part of the thesis describes essential theoretical knowledge about the anatomy of the axial skeleton, computer tomography, image processing methods and about the detection and segmentation issues. The second part contains the algorithms realisation description, the evaluation and the discussion of the results. Applications of the algorithms in CAD systems is described at the end. The application of all of the points is done in the programming software Matlab.
|
|
|
Segmentation of the cord canal and intervertebral discs in MRI data
Koban, Martin ; Odstrčilík, Jan (referee) ; Jakubíček, Roman (advisor)
The concern of this thesis is development of the method for the spinal canal and intervertebral discs segmentation in volume MRI data. The primary aim is to achieve the highest possible level of automation and accuracy allowing for reliable quantitative evaluation of the results. The algorithm is based on the random walk model in combination with a specific active contour method formulated through level set concept. The proposed approach is tested using a database of three-dimensional T2-weighted MR images, which also contains referential manual segmentation of intervertebral discs.
|
|
|
Convolutional neural networks for identification of axial 2D slices in CT data
Vavřinová, Pavlína ; Harabiš, Vratislav (referee) ; Jakubíček, Roman (advisor)
This thesis deals with the classification of axial 2D slices in CT patient’s data into six categories. The sphere of convolutional neural networks was used for this purpose. For a better understanding of this issue, the basics of neural networks and then the principles of deep learning including convolutional neural networks are explained at first. The AlexNet network was specifically selected for the intention of this identification, and it was tested on the created data set after being adaptated. The overall classification success rate was 86% ,after the final adjustments, a slight improvement was achieved and the identification success rate was 87%.
|
|
|
Evaluation of eye-blinking artifact effect on fusion result of simultaneous EEG-fMRI data
Dobiš, Lukáš ; Jakubíček, Roman (referee) ; Labounek, René (advisor)
This thesis sets a theoretical framework about simultaneous EEG-fMRI fusion. The work contains a description of basic principles of acquisition, their individual artifact types and preprocessing techniques for each type of data. Thesis mainly deals with suppression of eye blink artifacts in EEG data, by the method of independent component analysis. The following part explains the technique of simultaneous EEG-fMRI fusion in a general linear model and the creation of activation maps of statistically important correlations. This chapter is concluded with a description of methodology needed for result analysis. Finally, the used data are described, and a solution is proposed and applied in process of EEG preprocessing with artifact suppression, data fusion and result evaluation in MATLAB environment. Evaulation results showed that eye blink artifact influences the fusion result computed from relative power values more then that constructed via absolute power values. Tested method didnt supress eye blink artifact completely.
|
|
|
Head pose estimation via stereoscopic reconstruction
Hříbková, Veronika ; Jakubíček, Roman (referee) ; Kolář, Radim (advisor)
The thesis deals with head pose estimation in stereo data. The theoretical part provides the basis for understanding the geometry of the camera, its parameters and the method of calibration. The following describes the principles of stereo analysis and creating of disparity maps. In the research section, the methods used for head pose modelling are presented and an analysis of selected published articles is given. In the course of the master’s thesis, a system of two cameras for stereoscopic acquisition of motion of the head was designed and several measurements were carried out. The obtained data was prepared for creation of disparity maps and further processing. Based on the detection of facial features, in particular the inner and outer corners of the eyes and corners of the mouth, and their correspondences, a simple geometric model in shape of triangle was created to illustrate the inclination of the facial plane in space. By computing the angle of inclination in three axes, the current head pose is obtained. Motion is modelled by tracking detected points during video sequences.
|
|
|
Detection of specific anatomical structures in CT data via convolutional neural networks
Kozlová, Dominika ; Jan, Jiří (referee) ; Jakubíček, Roman (advisor)
This thesis deals with the issue of detection of anatomical structures in medical images using convolutional neural networks (CNN). At first there are described methods of machine learning, convolutional neural networks and selected methods for detection using CNN. In this work was created a database of annotated CT images of ten anatomical structures (head, heart, aorta, left and right lung, spine, liver, left and right kidney, spleen). A method for detecting these structures was designed, that contains two approaches of region proposals from image, CNN and postprocessing to obtain the detection result. The designed algorithm was implemented in the Python programming language using the TensorFlow library. Obtained results of validation of the network and the detection results are presented and discussed in the last chapter.
|
|
|
Segmentation of cortical parts of vertebrae
Janštová, Michaela ; Kolář, Radim (referee) ; Jakubíček, Roman (advisor)
This thesis deals with a segmentation of cortical parts of vertebrae from CT image datas in programming software called MATALB. Issues about segmentation techniques are described, especially „level-set” method and its modification DRLSE. This method was chosen because of informations from articles published in spcialized publications and also thanks to its plentiful usage and satisfactory results. In the end of this paper is designed method tested on real CT datas.
|
|
|
3D Slicer Extension for Tomographic Images Segmentation
Chalupa, Daniel ; Jakubíček, Roman (referee) ; Mikulka, Jan (advisor)
This work explores machine learning as a tool for medical images' classification. A literary research is contained concerning both classical and modern approaches to image segmentation. The main purpose of this work is to design and implement an extension for the 3D Slicer platform. The extension uses machine learning to classify images using set parameters. The extension is tested on tomographic images obtained by nuclear magnetic resonance and observes the accuracy of the classification and usability in practice.
|
|
|
Vertebra detection and identification in CT oncological data
Věžníková, Romana ; Harabiš, Vratislav (referee) ; Jakubíček, Roman (advisor)
Automated spine or vertebra detection and segmentation from CT images is a difficult task for several reasons. One of the reasons is unclear vertebra boundaries and indistinct boundaries between vertebra. Next reason is artifacts in images and high degree of anatomical complexity. This paper describes the design and implementation of vertebra detection and classification in CT images of cancer patients, which adds to the complexity because some of vertebrae are deformed. For the vertebra segmentation, the Otsu’s method is used. Vertebra detection is based on search of borders between individual vertebra in sagittal planes. Decision trees or the generalized Hough transform is applied for the identification whereas the vertebra searching is based on similarity between each vertebra model shape and planes of CT scans.
|
guest ::
