|
|
Advanced retinal vessel segmentation methods in colour fundus images
Svoboda, Ondřej ; Jan, Jiří (referee) ; Odstrčilík, Jan (advisor)
Segmentation of vasculature tree is an important step of the process of image processing. There are many methods of automatic blood vessel segmentation. These methods are based on matched filters, pattern recognition or image classification. Use of automatic retinal image processing greatly simplifies and accelerates retinal images diagnosis. The aim of the automatic image segmentation algorithms is thresholding. This work primarily deals with retinal image thresholding. We discuss a few works using local and global image thresholding and supervised image classification to segmentation of blood tree from retinal images. Subsequently is to set of results from two different methods used image classification and discuss effectiveness of the vessel segmentation. Use image classification instead of global thresholding changed statistics of first method on healthy part of HRF. Sensitivity and accuracy decreased to 62,32 %, respectively 94,99 %. Specificity increased to 95,75 %. Second method achieved sensitivity 69.24 %, specificity 98.86% and 95.29 % accuracy. Combining the results of both methods achieved sensitivity up to72.48%, specificity to 98.59% and the accuracy to 95.75%. This confirmed the assumption that the classifier will achieve better results. At the same time, was shown that extend the feature vector combining the results from both methods have increased sensitivity, specificity and accuracy.
|
|
|
Software for automatic data extraction in analysis of brain connectivity
Bujnošková, Eva ; Schwarz, Daniel (referee) ; Jan, Jiří (advisor)
The brain; complex system people want to know about but still they are at the beginning of understanding it. There has been a lot of neuroimaging systems since developement of modern technologies and magnetic resonance imaging is one of them. In last days it isn't enough to examine only structural character of brain, the scientists are dealing with functional states more and more; the functional magnetic resonance imaging is perfectly good tool for this. There is a big amount of researches concerning individual brain regions but also a lot of them dealing with communication across the brain to clear up the causes of human behavior and functional failures. This thesis introduces the brain connectivity exploration, it uses the parcellation by anatomical atlases and it tries to use the knowledge of graph theory as one of the options to determine relations between brain centres and regions. The thesis introduces the software created for extraction of connectivity matrix resulting in graph processing and visualization.
|
|
|
Mapping of motion artefact in fMRI
Nováková, Marie ; Kremláček, Jan (referee) ; Jan, Jiří (advisor)
This thesis summarizes a theory of magnetic resonance and the method of functional magnetic resonance. It is focused on the influence of motion artifacts and image preprocessing methods, especially realign. It deals with the possibility of using movement parameters obtained in the process of alignment of functional scans to create maps that show the expression of motion artifacts. In this thesis, three different methods were designed, implemented a tested. These methods lead to the creation of probability, power and statistical group maps showing areas typically affected by movement artifacts.
|
|
|
Comparison of advanced analysis of fMRI data from oddball experiment
Fajkus, Jiří ; Jan, Jiří (referee) ; Provazník, Ivo (advisor)
This master´s thesis deals with processing and analysis of data, acquired from experimental examination performed with functional magnetic resonance imaging technique. It is an oddball type experimental task and its goal is an examination of cognitive functions of the subject. The principles of functional magnetic resonance imaging, possibilities of experimental design, processing of acquired data, modeling of a response of organism and statistical analysis are described in this work. Furthermore, particular parts of preprocessing and analysis are carried out using real data set from experiment. The main goal of this work is suggestion and realization of model, which enables advanced categorization of stimuli, considering the type of previous rare stimulus and the number of frequent stimuli within them. With its in-depth categorization, this model enables detail exploration of cerebral processes, associated mainly with attention, memory, expectancy or cognitive closure. The second point of that work is an evaluation of models of hemodynamic response, which are applied in statistical analysis of data from fMRI experiment. Comparison of basis functions, the models of hemodynamic response to experimental stimulation used for general linear model, is performed in this work. The result of this comparison is an evaluation of detection efficiency of activated voxels, false positivity rate and computational and user difficulty.
|
|
|
A comparison of effective and functional connectivity methods in fMRI
Gajdoš, Martin ; Schwarz, Daniel (referee) ; Jan, Jiří (advisor)
Functional magnetic resonance imaging (fMRI) is recent important method, used in neuroimaging. The aim of this thesis is to develop software tool for comparison of two methods for functional and effective connectivity estimation. In this thesis are described the basics of magnetic resonance imaging, fMRI, basic terms of fMRI experiments and generally are described methods of functional and effective connectivity. Then are more detailed mentioned methods of dynamic causal modeling (DCM), Granger causal modeling (GCM) and independent component analysis (ICA). Practical implementation of DCM in toolbox SMP and ICA in toolbox GIFT is also mentioned. In purpose to describe behavior of DCM and GCM in dependence on several parameters are performed Monte Carlo simulations. Then the concept and realization of software tool for simulating connectivity and comparison of DCM and GCM are described. Finally results of DCM and GCM comparison and results of Monte Carlo simulations are discussed.
|
|
|
Software for patients and brain regions selection suitable for analysis of connectivity in fMRI
Slavíček, Tomáš ; Lamoš, Martin (referee) ; Jan, Jiří (advisor)
The aim of this thesis is to create an exploratory tool for functional magnetic resonance imaging data, which allows quickly and easily making a selection of persons and areas suitable for group analysis of connectivity. In the first chapters of this work is mentioned history of brain research and comparison of methods used in functional imaging. Next they are discussed the theoretical basis of fMRI methods, such as the formation of BOLD signal, acquisition parameters of MRI images and methods for designing experiments. The following chapter describes in detail the analysis of recorded data from the pre-processing to the interpretation of results. The last chapter of the first part describes problems of group analysis in SPM8 software. The second half of this work is dedicated to the description of developed program from data input to saving the results, including detailed descriptions of key features. In conclusion, there is a chapter characterizing the application of developed program on real data from clinical studies, including the results and evaluation of the usability of program. The program will mainly be used in neuroscience research.
|
|
|
Restoration of optical coherence tomography image data
Smékal, Ondřej ; Odstrčilík, Jan (referee) ; Jan, Jiří (advisor)
Restoration of image data has become an essential part of the processing of medical images obtained by any system. The same applies in the case of optical coherence tomography. The aim of this work is to study the first restoration methods. Second, the description of the data representation from optical coherence tomography and subsequent discussions that restoration methods based on deconvolution would potentially find application in processing of Optical coherence tomography. Finally, the third to create a program solution of the OCT data restoration process in MATLAB environment and followed by discussion of effectiveness of the presented solutions.
|
|
|
Textural Analysis of Nerve Fibre Layer in Retinal Images
Novotný, Adam ; Jan, Jiří (referee) ; Odstrčilík, Jan (advisor)
This work describes completely new approach to detection of retinal nerve fibre layer (RNFL) loss in colour fundus images. Such RNFL losses indicate eye glaucoma illness and an early diagnosis of RNFL changes is very important for successful treatment. Method is presented with the purpose of supporting glaucoma diagnosis in ophthalmology. The proposed textural analysis method utilizes local binary patterns (LBP). This approach is characterized especially by computational simplicity and insensitivity to monotonic changes of illumination. Image histograms of LBP distributions are used to gain several textural features aimed to classify healthy or glaucomatous tissue of the retina. The method was experimentally tested using fundus images of glaucomatous patients with focal RNFL loss. The results show that the proposed method can be used in order to supporting diagnosis of glaucoma with satisfactory efficiency.
|
|
|
Texture analysis of retinal images oriented towards detection of neronal fibre layer
Gazárek, Jiří ; Jiřík, Radovan (referee) ; Jan, Jiří (advisor)
The thesis is focused on detection of local disappearance of the neural layer on retina in fundus-camera images. The first chapter describes the human eye physiology, the glaucoma disease and the analyzed data. The second chapter compares four different approaches that should enable automatic detection of a possible damage to the retinal neural layer. These four approaches have been tested and evaluated; three of them showed an acceptable correlation with the medical expert conclusions – the directional spectral approach, the edge based approach and the difference local brightness. The last approch via local co-occurrence matrices has not turned out to be informative with the respect to the issue concerned. Then a program for the automatic detection of the nerve fibre layer loss areas has been designed, realized and evaluated. This task is solved in the last chapter. A relatively good agreement between the medical expert conclusions and the conclusions detected automatically by this program has been reached.
|
|
|
Analysis of Colour Retinal Images Aimed at Segmentation of Vessel Structures
Odstrčilík, Jan ; Jiřík, Radovan (referee) ; Jan, Jiří (advisor)
Segmentation of vessel structure is an important phase in analysis of retinal images. The resulting vessel system description may be important for diagnostic of many eye and cardiovascular diseases. A method for automatic segmentation of the vessel structure in colour retinal images is presented in the thesis. The method utilises 2D matched filtering to detect presence of short linear vessel sections of a particular thickness and orientation. The approach correlates the local image areas with a 2D masks based on a typical brightness profile perpendicular to vessels of a particular width. Three different approximated profiles are used and corresponding matched filters are designed for: thin, medium and thick vessels. The evaluation of typical vessel profiles and filter design are described in chapter 3 and chapter 4. The parametric images obtained by convolution of the image with the masks are then thresholded in order to obtain binary representation of vessel structure. The three binary representations are consequently combined to provide the best available rough vessel map, which is finalised by complementing the obviously missing vessel sections and cleaning the disconnected fractional artefacts. The thresholding algorithm and final steps of processing are mentioned in chapter 5 and chapter 6. The method has been implemented by computer and the program for automatic vessel segmentation has been developed using database of real retinal images. The efficiency of the method has been finally evaluated on images from the standard database DRIVE.
|
guest ::
RSS feed.