|
|
Face recognition
Maňkoš, Richard ; Mézl, Martin (referee) ; Kolář, Radim (advisor)
This diploma thesis deals with face recognition in digital pictures. The first part describes biometry and, shortly, characterizes biometrical methods which are the most oftenly used. In the second part is described the approach of face recognition in a picture. Specifically, it is described the method for face detection - Viola-Jones and method for face recognition - PCA, which will be implemented in Matlab. The last part, which is practical, describes the scheme for video-sequence recording, implementation of the PCA method in Matlab and discussion of the achieved results.
|
|
|
Quantification of Perfusion using MRI ASL Methods
Kolajová, Martina ; Mézl, Martin (referee) ; Jiřík, Radovan (advisor)
This bachelor thesis deals with quantification of perfusion using a contrast-free MRI technique called „Arterial Spin Labeling“ (ASL). Firstly, the physical principle of magnetic resonance imaging and the concept of perfusion will be described for better understanding of the subject. The next chapters will focus on the ASL method itself and perfusion quantification using mathematical models, namely T1 perfusion and Standard. The algorithm for preprocessing and analysis of selected datasets, designed in the thesis, is implemented using the program MATLAB. Program BASIL was used for further processing. The outputs in the form of perfusion and ATT maps and model fitting processes using various models from both programs are compared and evaluated.
|
|
| |
|
|
Methods for sleep spindles detection from EEG records
Matoušek, Šimon ; Mézl, Martin (referee) ; Králík, Martin (advisor)
This bachelor work focuses on the detection of sleep spindles in EEG signals. The introductory chapter deals with the EEG signal, describes its components and describes the signal recording process. Explains the term sleep spindle and clarifies polysomnography. In the following chapter, some findings concerning studies that examined and practically used individual methods of sleep spindle detection are summarized in the form of research. The practical part of the work is focused on some sleep spindle detectors. At the end of the work is a comparison of the success of these detectors in comparison with other, previously performed studies. The highest success was achieved with the detector based on signal envelope calculation, where the sensitivity was 56.00 \% and the specificity 55.19 %, and also with the detector using wavelet transforms, where the sensitivity was 81.22 % and the specificity 46.15 %
|
|
|
Image registration using evolutionary algorithms
Kubalová, Eva ; Kozumplík, Jiří (referee) ; Mézl, Martin (advisor)
The diploma thesis deals with the image registration using evolutionary algorithms from metaheuristic optimization techniques which are considered recent and widely used. The frst part of the thesis contains theoretical description of components in image registration and later focuses on the ultrasound images. In that part, the thesis explains chosen evolutionary algorithms. Three optimization methods have been implemented, in particular genetic algorithm, particle swarm optimization and frey algorithm. Chosen similarity metrics for optimization are sum of squared dierences, cosine similarity and correlation coefcient. The main part of thesis includes testing of proposed methods with the evaluation of obtained results. These parameters are later used for optimization of real ultrasound sequences obtained by the contrast imaging.
|
|
|
Segmentation of optic disc in retinal image data
Juráček, Radek ; Mézl, Martin (referee) ; Odstrčilík, Jan (advisor)
This bachelor thesis is focused on the algorithm of automatic detection of the optical disk in retinal images. It briefly describes the anatomy of the human eye and the principles of scanning the ocular background. The following describes the optical disk segmentation methods. Selected methods are implemented in MATLAB and optimized using a genetic algorithm. A total of five methods were introduced and optimized on the HRF dataset and two experimental datasets.
|
|
|
Advanced Data Mining in Cardiology
Mézl, Martin ; Provazník, Ivo (referee) ; Sekora, Jiří (advisor)
The aim of this master´s thesis is to analyse and search unusual dependencies in database of patients from Internal Cardiology Clinic Faculty Hospital Brno. The part of the work is theoretical overview of common data mining methods used in medicine, especially decision trees, naive Bayesian classifier, artificial neural networks and association rules. Looking for unusual dependencies between atributes is realized by association rules and naive Bayesian classifier. The output of this work is a complex system for Knowledge discovery in databases process for any data set. This work was realized with collaboration of Internal Cardiology Clinic Faculty Hospital Brno. All programs were made in Matlab 7.0.1.
|
|
|
Image segmentation of spinal disc in medical imaging
Meloun, Jan ; Nemček, Jakub (referee) ; Mézl, Martin (advisor)
The thesis is focused on the segmentation of the intervertebral disc in the image data.The introduction deals with the issue of the spine, the herniation of the intervertebraldisc. It also deals with imaging modalities, especially computed tomography and mag-netic resonance imaging. The practical part describes the image data segmentation andthe implementation of three of the published segmentation methods.
|
|
| |
|
|
K-complex detection in sleep EEG
Bjelová, Martina ; Mézl, Martin (referee) ; Králík, Martin (advisor)
This paper addresses detecting of K-complexes in sleeping EEG records. Polysomnography is the method, which is used for diagnostic and following therapy of many sleep disorders. For identifnging of sleep stages it is fundamental to know graphoelements, in which they are situate. K-complex is important indicator of second sleep stange and hence is essencial to know to detect this pattern. In this paper we focus on design and implementation of more algorithms for detection of these patterns with various characteristics. Among the proposed methods, the wavelet transform method was best evaluated. Performance of this detection reached values the average senzitivity 63,83 % and average positive predictive value 44,07 %.
|
guest ::
