|
|
Feature extraction from image data
Uher, Václav ; Beneš, Radek (referee) ; Burget, Radim (advisor)
Image processing is one area of signal analysis. This thesis is involved in feature extraction from image data and its implementation using Java programming language. The main contribution of this thesis lies in develop features extractors and their implementation in the program RapidMiner. The result is a robust tool for image analysis. The functionality of each operator is tested on mammogram images. A function model was developed for the removal of artifacts from the mammography images. The success rate of removal is comparable with other similar works. Furthermore, learning algorithms were compared on example detection of ventricle in ultrasound image.
|
|
|
Methods for biomedical image signal segmentation in Java
Románek, Jakub ; Smékal, Zdeněk (referee) ; Šmirg, Ondřej (advisor)
This thesis contains two main parts, theoretical and implementation. In the theoretical part, there are described the different segmentation methods. Mainly it is about description method Level Set. The aim of practical part was to create a java module for segmentation of biomedical images using Level Set methods. The work solves example of a simple GUI for the display of results.
|
|
|
Image database system for glaucoma diagnosis support
Peter, Roman ; Roleček, Jiří (referee) ; Kolář, Radim (advisor)
Tato práce popisuje přehled standardních a pokročilých metod používaných k diagnose glaukomu v ranném stádiu. Na základě teoretických poznatků je implementován internetově orientovaný informační systém pro oční lékaře, který má tři hlavní cíle. Prvním cílem je možnost sdílení osobních dat konkrétního pacienta bez nutnosti posílat tato data internetem. Druhým cílem je vytvořit účet pacienta založený na kompletním očním vyšetření. Posledním cílem je aplikovat algoritmus pro registraci intenzitního a barevného fundus obrazu a na jeho základě vytvořit internetově orientovanou tři-dimenzionální vizualizaci optického disku. Tato práce je součásti DAAD spolupráce mezi Ústavem Biomedicínského Inženýrství, Vysokého Učení Technického v Brně, Oční klinikou v Erlangenu a Ústavem Informačních Technologií, Friedrich-Alexander University, Erlangen-Nurnberg.
|
|
|
Software FLIM system with pulse white light laser in confocal microscopy
Grund, Pavel ; Odstrčilík, Jan (referee) ; Čmiel, Vratislav (advisor)
The theoretical part of this master's thesis is focused on research of confocal microscopy and FLIM method. There are a principles and types of confocal microscopy and the use of broad-spectrum laser as a basic light source of these microscopes. It gives what the FLIM method and its use not only in cell biology. The practical part thesis includes the acquisition of three sets of fluorescence intensity images with use of applications tunable pulsed laser, function TimeGate and detection of hybrid detectors. For practical elaboration of this thesis is in the software Fiji created a plugin, which is the source code in the Java programming language. The types of plugins and their uses are described in the third chapter of the thesis. This plugin including the graphical user interface in the form of the dialog box, proceses the fluorescence intensity images and creates a graphical representation of data showing the fluorescence lifetime, so called pseudocolor map.
|
|
|
Automatic medical image segmentation using Level Set algorithm
Yerpeissov, Serik ; Burget, Radim (referee) ; Uher, Václav (advisor)
This thesis contains two main parts, theoretical and implementation. In the theoretical part, there are described the different segmentation methods. Mainly it is about description method Level Set. The aim of practical part is creation of java module for segmentation of medical data by using Level Set Methods. The work solves example of GUI for the display of results. The results should be tested
|
|
|
Altered morphology of white and grey matter in patients with Alzheimer disease and Schizophrenia on MRI
Lahutsina, Anastasiya ; Zach, Petr (advisor) ; Horáček, Jiří (referee) ; Němcová, Veronika (referee)
Cortical folding of the anterior cingulate cortex (ACC), particularly the cingulate (CS) and the paracingulate (PCS) sulci, represents a neurodevelopmental marker. Deviations in in utero development in schizophrenia can be traced using CS and PCS morphometry. In the present study, we measured the length of CS, PCS, and their segments on T1 MRI scans in 93 patients with first episode schizophrenia and 42 healthy controls. Besides the length, the frequency and the left-right asymmetry of CS/PCS were compared in patients and controls. Distribution of the CS and PCS morphotypes in patients was different from controls. Parcellated sulcal pattern CS3a in the left hemisphere was longer in patients (53.8 ± 25.7 mm vs. 32.7 ± 19.4 mm in controls, p < 0.05), while in CS3c it was reversed-longer in controls (52.5 ± 22.5 mm as opposed to 36.2 ± 12.9 mm, n.s. in patients). Non parcellated PCS in the right hemisphere were longer in patients compared to controls (19.4 ± 10.2 mm vs. 12.1 ± 12.4 mm, p < 0.001). Therefore, concurrent presence of PCS1 and CS1 in the left hemisphere and to some extent in the right hemisphere may be suggestive of a higher probability of schizophrenia. Measurement of an hippocampal area or volume is useful in clinical practice as a supportive aid for diagnosis of Alzheimer's disease....
|
|
|
Automatic medical image segmentation using Level Set algorithm
Yerpeissov, Serik ; Burget, Radim (referee) ; Uher, Václav (advisor)
This thesis contains two main parts, theoretical and implementation. In the theoretical part, there are described the different segmentation methods. Mainly it is about description method Level Set. The aim of practical part is creation of java module for segmentation of medical data by using Level Set Methods. The work solves example of GUI for the display of results. The results should be tested
|
|
|
Software FLIM system with pulse white light laser in confocal microscopy
Grund, Pavel ; Odstrčilík, Jan (referee) ; Čmiel, Vratislav (advisor)
The theoretical part of this master's thesis is focused on research of confocal microscopy and FLIM method. There are a principles and types of confocal microscopy and the use of broad-spectrum laser as a basic light source of these microscopes. It gives what the FLIM method and its use not only in cell biology. The practical part thesis includes the acquisition of three sets of fluorescence intensity images with use of applications tunable pulsed laser, function TimeGate and detection of hybrid detectors. For practical elaboration of this thesis is in the software Fiji created a plugin, which is the source code in the Java programming language. The types of plugins and their uses are described in the third chapter of the thesis. This plugin including the graphical user interface in the form of the dialog box, proceses the fluorescence intensity images and creates a graphical representation of data showing the fluorescence lifetime, so called pseudocolor map.
|
|
|
Methods for biomedical image signal segmentation in Java
Románek, Jakub ; Smékal, Zdeněk (referee) ; Šmirg, Ondřej (advisor)
This thesis contains two main parts, theoretical and implementation. In the theoretical part, there are described the different segmentation methods. Mainly it is about description method Level Set. The aim of practical part was to create a java module for segmentation of biomedical images using Level Set methods. The work solves example of a simple GUI for the display of results.
|
|
|
Feature extraction from image data
Uher, Václav ; Beneš, Radek (referee) ; Burget, Radim (advisor)
Image processing is one area of signal analysis. This thesis is involved in feature extraction from image data and its implementation using Java programming language. The main contribution of this thesis lies in develop features extractors and their implementation in the program RapidMiner. The result is a robust tool for image analysis. The functionality of each operator is tested on mammogram images. A function model was developed for the removal of artifacts from the mammography images. The success rate of removal is comparable with other similar works. Furthermore, learning algorithms were compared on example detection of ventricle in ultrasound image.
|
guest ::
