|
|
Defects of pre-mRNA splicing causing retitinis pigmentosa
Pakhomova, Yelyzaveta ; Staněk, David (advisor) ; Vaňková Hausnerová, Viola (referee)
Retinitis pigmentosa is a genetic disorder affecting the retina. The progression of the disease leads to vision loss. This thesis concentrates on the causation of autosomal dominant retinitis pigmentosa. More specifically, the second biggest responsible mutation group is outlined. The above mentioned gene-mutations group is responsible for the formation of mutant variants of their corresponding splicing proteins. These proteins and consequences of their mutations are reviewed and presented in the thesis. The outline of mutation impact on the retina is presented for each mutated protein. The proteins in question are: PRPF8, PRPF31, PRPF3, PRPF4, PRPF6, SNRNP200, DHX38 (an exemption causing an autosomal recessive retinitis pigmentosa), PAP-1, CWC27 (an exemption causing an autosomal recessive retinitis pigmentosa). The literature review allowed the thesis to conclude that splicing proteins are highly likely to play a critical role in retina's health. In addition, some other noteworthy findings are briefly presented. For example, findings regarding lack of data about some of the mutations. Another example of such finding is that it still remains unknown why these mutations cause such a tissue-specific phenotype. Key words: splicing, retinitis pigmentosa, snRNP, retina, autosomal dominant retinitis...
|
|
|
Segmentation of arterial wall in high resolution retinal images
Polachová, Natálie ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
This thesis focuses on automatic segmentation of retinal arterial walls in images acquired using adaptive optics. Adaptive optics is a non-invasive imaging method that provides high lateral resolution and allows detailed observation of retinal microstructures, including arterial walls. This technology is crucial for early diagnosis of serious diseases such as arterial hypertension and diabetic retinopathy. The main objective of this work was to detect the arterial lumen and segment its walls. Morphological and filtration techniques were used for lumen detection. For arterial wall segmentation, brightness profiles along the detected lumen were analyzed and active contour and spline methods were used. The results show that the active contour segmentation method improves the accuracy of arterial wall detection, especially in high-contrast regions. This thesis summarizes the findings and proposes improvements in the detection of the inner side of the arterial wall, which reduces the segmentation success rate in this work.
|
|
|
Registration of retinal image data using evolutionary algorithms
Matoušek, Šimon ; Mézl, Martin (referee) ; Odstrčilík, Jan (advisor)
This thesis deals with the registration of retinal images. Firstly, the introductory chapter summarizes the theory related to evolutionary algorithms and their practical application in the form of a literature search, then different approaches of retinal image registration and image transformation are described. At the same time, the anatomy of the eye as a key organ for visual perception is briefly introduced. In the practical part of the thesis, the method of registration of static and then dynamic retinal images is implemented using MATLAB software environment, and finally the results of different approaches are compared.
|
|
|
Detection of Diseases Caused by Diabetes in Retinal Images
Zapletal, Michal ; Semerád, Lukáš (referee) ; Kavetskyi, Andrii (advisor)
The goal of this thesis is to design and implement an algorithm for detecting exudates and microaneurysms in colored retinal images. These diseases are the first signs of diabetic retinopathy and early detection is crucial. The proposed algorithm begins with preprocessing, where excess background is removed, contrast is enhanced using CLAHE and histogram stretching, and noise filtering is applied. Optic disc localization is based on iterative background removal and row and column variances. Exudates detection is performed based on gamma correction, thresholding and optic disc removal. Microaneurysm detection is based on morphological operations, hit-or-miss transformation and principal component analysis (PCA). The algorithm was tested on 4 datasets with accuracy 73,1 % for exudates and 73,3 % for microaneurysms. The resulting program could assist in automatic disease detection, which could potentially save time for doctors.
|
|
|
Semantic Segmentation of Pathologies in Retinal Images
Čabala, Roman ; Orság, Filip (referee) ; Kavetskyi, Andrii (advisor)
The thesis aimed to segment pathology visible in the retina images, such as exudates, hemorrhages, and microaneurysms. For that, two well known deep neural networks, named U-Net and SegFormer, were trained. To test the performance of the models, one publicly available dataset was used, named IDRiD. Obtained results were reported after analyzing different factors which affected the performance of the models U-Net and Segformer.
|
|
|
Image Registration - Application in ophthalmology and ultrasonography
Harabiš, Vratislav ; Matula,, Petr (referee) ; Zemčík, Pavel (referee) ; Kolář, Radim (advisor)
Image registration is widely used in clinical practice. However image registration and its~evaluation is still challenging especially with regards to new possibilities of various modalities. One of these areas is contrast-enhanced ultrasound imaging. The time-dependent image contrast, low signal-to-noise ratio and specific speckle pattern make preprocessing and image registration difficult. In this thesis a method for registration of images in ultrasound contrast-enhanced sequences is proposed. The method is based on automatic fragmentation into image subsequences in which the images with similar characteristics are registered. The new evaluation method based on comparison of perfusion model is proposed. Registration and evaluation method was tested on a flow phantom and real patient data and compared with a standard methods proposed i literature. The second part of this thesis contains examples of application of image registration in~ophthalmology and proposition for its improvement. In this area the image registration methods are widely used, especially landmark based image registration method. In this thesis methods for landmark detection and its correspondence estimation are proposed.
|
|
|
Application for Recognition of Human Eye Retina
Drozd, Radek ; Hájek, Josef (referee) ; Drahanský, Martin (advisor)
The blood vessels layout in a human eye retina is unique for every person in the world, so it is one of important biometric characteristics. Processing of colour retina image may be a part of an intended biometric system. There is an algorithm for automatic blood vessels detection, optic disc and macula localisation, finding of bifurcation points and saving those as a biometric template presented in this bachelor's thesis. C++ programming language and OpenCV library were used for implementation. The application was tested on a set of colour retina images, taken by fundus camera. The final application is supposed to run on a digital signal processor, developed by Texas Instruments. The thesis gives the introduction into biometrics, signal processing and human eye anatomy.
|
|
|
Biometry based on retinal videosequences
Oweis, Kamil ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
The biometric methods are the most advanced methods for recognition and verification of person identity. These methods are quite fast, safe and applicable in different situations. In this thesis is used a set of retinal scans taken with a video-ophtalmoscope. These pictures are further modified for next processing, first of all by convertion into black-andwhite binary image, in some cases was after that used a binary matrix for description of image. Afterwards was suggested comparison method of images from the database with reference image of the retina: method of overlap and shift. It was tested a set of blackand-white and then also grey images. All method calculations was realized in program Matlab of which outcome was determination of the most congruent image with reference image and evaluation of overall program accuracy.
|
|
|
Criterial function for retinal image registration
Horáková, Pavla ; Kolář, Radim (referee) ; Harabiš, Vratislav (advisor)
This bachelor‘s thesis is focused on comparison criterial functions, which are usually used in registration of retinal images. Criterial functions reflect degree of images’ sameness thereby they influence proper registration considerably. The aim of this thesis is to discover criterial function, which is the best for image registration of retina. Registration of images is very difficult; it is important to get some informations such as an uniqueness or a shape of function.
|
|
|
Retinal Blood Vessel Segmentation
Nemčeková, Barbora ; Drahanský, Martin (referee) ; Kavetskyi, Andrii (advisor)
The retina is an important part of the human eye. Incident light is processed here and moreover, it plays an essential role in diagnosing various diseases. Its early diagnostics can prevent serious consequences, such as blindness. The most common retinal diseases include diabetic retinopathy, as a consequence of diabetes, and age-related macular degeneration. Automatic retinal vessels segmentation facilitates and speeds up the work of an ophthalmologist. This work focuses on retinal blood vessels segmentation and its further classification into thin and thick vessels. The proposed algorithm is based on morphological operations, k-means clustering, and Frangi's algorithm. Evaluation of the proposed method was performed on two publicly available datasets - Drive and HRF. The results obtained represent 69,89 % for sensitivity, 91,55 % for specificity, and 88,63 % for accuracy. Division of the vessels shows, that on average 21,50 % vessels pixels belong to thick vessels and the rest 78,50 % belong to thin vessels.
|
guest ::
