|
|
Deep learning methods for vessel and optic disc segmentation in ophthalmologic sequences
Rozhoňová, Andrea ; Odstrčilík, Jan (referee) ; Hesko, Branislav (advisor)
The aim of the following thesis was to study the issue of optical disc and retinal vessels segmentation in ophthalmologic sequences. The theoretical part of the thesis summarizes the principles of different approaches in the field of deep learning, which are used in connection with the given issue. Based on the theoretical part, methods for optical disk segmentation and retinal vessel segmentation based on the convolutional neural networks Linknet, PSPNet, Unet and MaskRCNN are proposed. The practical part of the thesis deals with the description of their implementation and subsequent evaluation.
|
|
|
Detection of optic disc from fundus camera sequences
Juráček, Radek ; Harabiš, Vratislav (referee) ; Hracho, Michal (advisor)
This semestral thesis deals with shape detection in images and application of it for detection of optic disk in sequence of fundus camera images. It describes major features of the human eye and retinal diseases. Further the thesis discusses some methods of image preprocessing, segmentation and generalised Hough transform for fundus images which is the main work content for optic disc localization. Practical part describes the proposed methods for detection of optical disk based on a circular Hough transformation and adapted filtering. Method of adaptad filtering achieved median of overlap 59,1%, method using fast Hough transform algorithm achieved median of overlap 24,95%, method using clasic Hough transform algorithm achieved median of overlap 29,95%.
|
|
|
Optic Disc Detection in retinal video-sequences
Černohorská, Lucie ; Odstrčilík, Jan (referee) ; Labounková, Ivana (advisor)
This semestral thesis deals with the detection of optical disc from retina images taken by experimental video – ophthalmoscope. There is briefly destribed anatomy of human eye, its illness and also overview of imaging and diagnostics methods of retinal. The thesis discussed several methods, which can be used for the detection of optical disc. The practical part of semestral thesis is focused on application of Hough transform on images from ophthalmoscope. The suggested algorithm is tested on 25 retinal sequences. The accuracy of detection of optical disc on still image is 71,10 %. The thesis deals with detection of OD movement and the accuracy of the detection is evaluated using a reference movement signal.
|
|
|
Optic nerve head segmentation in retinal image data
Nohel, Michal ; Drahanský, Martin (referee) ; Kolář, Radim (advisor)
This diploma thesis deals with the segmentation of the optic disc and cup in retinal image data. The theoretical part of the thesis describes the optic disc and cup and provides an overview of the current state of the art in using machine learning methods for their segmentation. Furthermore, the basic principles and blocks of convolutional neural networks are described. Convolutional neural networks U-Net and its modification nnU-Net were trained on the created databases. These models were tested and the results obtained were discussed and compared with selected published methods. Finally, the models were evaluated in terms of their potential for practical application.
|
|
|
Visualization and analysis of retinal pulsatile phenomena
Plavcová, Daniela ; Odstrčilík, Jan (referee) ; Kolář, Radim (advisor)
Retinal blood flow can be a reflection of the state of the vascular system, but it can also indicate changes in intracranial pressure. For that reason, video sequences are taken from the retina, which can be processed to examine this area. This work deals with the analysis and visualization of pulsations on the retina, representing the change in light attenuation during one hearth cycle, and the search for suitable parameters describing connections and differences between data, whether in the form of signals, images or video sequences.
|
|
|
Advanced registration of image sequences from video-ophthalmoscope
Dufková, Barbora ; Chmelík, Jiří (referee) ; Kolář, Radim (advisor)
This master's thesis deals with the issue of registration of ophthalmic video sequences. It describes basic geometric transformations that can be used for registration. The basic methods of image registration are also presented, from which the most suitable variant for this application is selected. This is then implemented using a script created in the MATLAB environment. The proposed method is further evaluated objectively using the brightness profile method, using mutual information and correlation, and using retinal vessel skeleton. The effect of polynomial transformation on registration and possible optimizations of the algorithm are discussed.
|
|
|
Detection of optic disc in retinal fundus image
Nohel, M. ; Kolář, R.
This paper presents a comparison of several image processing methods for optic disc detection in retinal images. The detection algorithms are tested on five publicly available retinal databases (total 2140 images). Thresholding, Hough transform, and matched filtering followed by postprocessing were tested for optic disc detection. It has been shown that matched filtering followed by peak detection achieves the best success rate (98.8 %).
|
|
|
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.
|
|
|
Deep learning methods for vessel and optic disc segmentation in ophthalmologic sequences
Rozhoňová, Andrea ; Odstrčilík, Jan (referee) ; Hesko, Branislav (advisor)
The aim of the following thesis was to study the issue of optical disc and retinal vessels segmentation in ophthalmologic sequences. The theoretical part of the thesis summarizes the principles of different approaches in the field of deep learning, which are used in connection with the given issue. Based on the theoretical part, methods for optical disk segmentation and retinal vessel segmentation based on the convolutional neural networks Linknet, PSPNet, Unet and MaskRCNN are proposed. The practical part of the thesis deals with the description of their implementation and subsequent evaluation.
|
|
|
Detection of optic disc from fundus camera sequences
Juráček, Radek ; Harabiš, Vratislav (referee) ; Hracho, Michal (advisor)
This bachelor thesis deals with optical disc detection in retinal images taken by an experimental video ophthalmoscope. At the beginning, it briefly describes the anatomy of the eyesight and retina, retinal disease and selected diagnostic methods of retinal imaging. There are described methods of detecting objects in the image, which are subsequently used in the practical part dealing with optical disc detection using Hough transformation and matched filtration. Proposed procedures were tested on a 100-image set. Hough Transformation Detection was successful in 64% of cases, matched filtration was successful in 44% of cases.
|
guest ::
