Yes / Detecting multiple ocular diseases in fundus images is crucial in ophthalmic diagnosis. This study introduces the Fundus-DeepNet system, an automated multi-label deep learning classification system designed to identify multiple ocular diseases by integrating feature representations from pairs of fundus images (e.g., left and right eyes). The study initiates with a comprehensive image pre-processing procedure, including circular border cropping, image resizing, contrast enhancement, noise removal, and data augmentation. Subsequently, discriminative deep feature representations are extracted using multiple deep learning blocks, namely the High-Resolution Network (HRNet) and Attention Block, which serve as feature descriptors. The SENet Block is then applied to further enhance the quality and robustness of feature representations from a pair of fundus images, ultimately consolidating them into a single feature representation. Finally, a sophisticated classification model, known as a Discriminative Restricted Boltzmann Machine (DRBM), is employed. By incorporating a Softmax layer, this DRBM is adept at generating a probability distribution that specifically identifies eight different ocular diseases. Extensive experiments were conducted on the challenging Ophthalmic Image Analysis-Ocular Disease Intelligent Recognition (OIA-ODIR) dataset, comprising diverse fundus images depicting eight different ocular diseases. The Fundus-DeepNet system demonstrated F1-scores, Kappa scores, AUC, and final scores of 88.56%, 88.92%, 99.76%, and 92.41% in the off-site test set, and 89.13%, 88.98%, 99.86%, and 92.66% in the on-site test set.In summary, the Fundus-DeepNet system exhibits outstanding proficiency in accurately detecting multiple ocular diseases, offering a promising solution for early diagnosis and treatment in ophthalmology. / European Union under the REFRESH – Research Excellence for Region Sustainability and High-tech Industries project number CZ.10.03.01/00/22_003/0000048 via the Operational Program Just Transition. The Ministry of Education, Youth, and Sports of the Czech Republic - Technical University of Ostrava, Czechia under Grants SP2023/039 and SP2023/042.

Submitted by Sandro Camargo (sandro.camargo@unipampa.edu.br) on 2015-05-09T18:40:27Z No. of bitstreams: 1 117110023.pdf: 2523085 bytes, checksum: 76eb3f9960e2b4f9df14435d3092b156 (MD5) / Made available in DSpace on 2015-05-09T18:40:27Z (GMT). No. of bitstreams: 1 117110023.pdf: 2523085 bytes, checksum: 76eb3f9960e2b4f9df14435d3092b156 (MD5) Previous issue date: 2013-12-13 / Este trabalho tem como objetivo a proposição de um novo método para a detecção automática de microaneurismas e hemorragias em imagens de fundo do olho. Essas lesões são consideradas o primeiro sinal de retinopatia diabética. A retinopatia diabética é uma doença originada pelo diabetes e é apontada com a principal causa de cegueira na população com idade ativa de trabalho. O método proposto é fundamentado em conceitos de morfologia matemática e consiste em eliminar os componentes da anatomia da retina até atingir o conjunto de lesões. Este método é formado por cinco etapas: a) pré-processamento; b) destaque das estruturas escuras; c) detecção dos vasos sanguíneos; d) eliminação dos vasos sanguíneos; e) eliminação da fóvea. A precisão do método foi testada num banco de dados público de imagens de fundo do olho, onde o mesmo obteve resultados satisfatórios e comparáveis aos demais métodos da literatura, reportando médias de sensitividade e especificidade de 87.69% e 92.44%, respectivamente. / This contribution presents an approach for automatic detection of microaneurysms and hemorrhages in fundus images. These lesions are considered the earliest signs of diabetic retinopathy. The diabetic retinopathy is a disease caused by diabetes and is considered as the major cause of blindness in working age population. The proposed method is based on mathematical morphology and consists in removing components of retinal anatomy to reach the lesions. This method consists of five steps: a) pre-processing; b) enhancement of low intensity structures; c) detection of blood vessels; d) elimination of blood vessels; e) elimination of the fovea. The accuracy of the method was tested on a public database of fundus images, where it achieved satisfactory results, comparable to other methods from the literature, reporting 87.69% and 92.44% of mean sensitivity and specificity, respectively.

Ocular fundus image can provide information on pathological changes caused by local ocular diseases and early signs of certain systemic diseases. Automated analysis and interpretation of fundus images has become a necessary and important diagnostic procedure in ophthalmology. Among the features in ocular fundus image are the optic disc, fovea (central vision area), lesions, and retinal vessels. These features are useful in revealing the states of diseases in the form of measurable abnormalities such as length of diameter, change in color, and degree of tortuosity in the vessels. In addition, retinal vessels can also serve as landmarks for image-guided laser treatment of choroidal neovascularization. Thus, reliable methods for blood vessel detection that preserve various vessel measurements are needed. In this paper, we will examine the pathological issues in the analysis of retinal vessels in digital fundus images and give a survey of current image processing methods for extracting vessels in retinal images with a view to categorize them and highlight their differences and similarities. We have also implemented two major approaches using matched filter and mathematical morphology respectively and compared their performances. Some prospective research directions are identified. / Singapore-MIT Alliance (SMA)

The goal of this thesis is to design and implement a program capable of automatically generating synthetic images of eye retinas. The generated images should be similar to those of real retinas, which are hard to obtain, so they could be used for development of various algorithms, which work with eye retina images in their place. This thesis describes anatomic properties of the eye retina, ways to take images of it and also usage of eye retina recognition in biometric and medicinal applications. Design of the program and the way in which individual parts of the retina are assembled together into the final image is also explained in this thesis. These individual parts are created using procedural textures in separate layers of the image. Next chapter of this work describes implementation details of the program. The conclusion then experimentally verifies suitability of the generated images for algorithmic processing.

Através das imagens de fundo do olho, os especialistas em oftalmologia podem detectar possíveis complicações relacionadas ao Diabetes como a diminuição ou até a perda da capacidade de visão. O Edema Macular Diabético (EMD) é uma das complicações que lideram os casos de danos à visão em pessoas em idade de trabalho. Sendo assim, esta tese apresenta métodos para automaticamente identificar os diferentes níveis de gravidade do Edema Macular Diabético visando auxiliar o especialista no diagnóstico dessa patologia. Como resultado final, propõe-se automaticamente e rapidamente identificar, a partir da imagem, se o paciente possui o EMD leve, moderado ou grave. Utilizando imagens de fundo do olho de um banco de dados livremente disponível na internet (ou seja, o DIARETDB1), o método proposto para a identificação automática do EMD obteve uma precisão de 94,29%. Alguns métodos intermediários necessários para a solução desse problema foram propostos e os resultados publicados na literatura científica. / Through color eye fundus images, the eye care specialists can detect possible complications related to diabetes as the vision impairment or vision loss. The Diabetic Macular Edema (DME) is the most common cause of vision damage in working-age people. Therefore, this thesis presents an approach to automatically identify the different levels of severity of diabetic macular edema aiming to assist the expert in the diagnosis of this pathology. As a final result, a methodology to automatically and quickly identify, from the eye fundus image, if a patient has the EMD mild, moderate or severe EMD is proposed. In a preliminary evaluation of our DME grading scheme using publicly available eye fundus images (i.e., DIARETDB1 image database), an accuracy of 94.29% was obtained. Some intermediate methods needed to solve this problem have been proposed and the results published in scientific literature.

Através das imagens de fundo do olho, os especialistas em oftalmologia podem detectar possíveis complicações relacionadas ao Diabetes como a diminuição ou até a perda da capacidade de visão. O Edema Macular Diabético (EMD) é uma das complicações que lideram os casos de danos à visão em pessoas em idade de trabalho. Sendo assim, esta tese apresenta métodos para automaticamente identificar os diferentes níveis de gravidade do Edema Macular Diabético visando auxiliar o especialista no diagnóstico dessa patologia. Como resultado final, propõe-se automaticamente e rapidamente identificar, a partir da imagem, se o paciente possui o EMD leve, moderado ou grave. Utilizando imagens de fundo do olho de um banco de dados livremente disponível na internet (ou seja, o DIARETDB1), o método proposto para a identificação automática do EMD obteve uma precisão de 94,29%. Alguns métodos intermediários necessários para a solução desse problema foram propostos e os resultados publicados na literatura científica. / Through color eye fundus images, the eye care specialists can detect possible complications related to diabetes as the vision impairment or vision loss. The Diabetic Macular Edema (DME) is the most common cause of vision damage in working-age people. Therefore, this thesis presents an approach to automatically identify the different levels of severity of diabetic macular edema aiming to assist the expert in the diagnosis of this pathology. As a final result, a methodology to automatically and quickly identify, from the eye fundus image, if a patient has the EMD mild, moderate or severe EMD is proposed. In a preliminary evaluation of our DME grading scheme using publicly available eye fundus images (i.e., DIARETDB1 image database), an accuracy of 94.29% was obtained. Some intermediate methods needed to solve this problem have been proposed and the results published in scientific literature.

Através das imagens de fundo do olho, os especialistas em oftalmologia podem detectar possíveis complicações relacionadas ao Diabetes como a diminuição ou até a perda da capacidade de visão. O Edema Macular Diabético (EMD) é uma das complicações que lideram os casos de danos à visão em pessoas em idade de trabalho. Sendo assim, esta tese apresenta métodos para automaticamente identificar os diferentes níveis de gravidade do Edema Macular Diabético visando auxiliar o especialista no diagnóstico dessa patologia. Como resultado final, propõe-se automaticamente e rapidamente identificar, a partir da imagem, se o paciente possui o EMD leve, moderado ou grave. Utilizando imagens de fundo do olho de um banco de dados livremente disponível na internet (ou seja, o DIARETDB1), o método proposto para a identificação automática do EMD obteve uma precisão de 94,29%. Alguns métodos intermediários necessários para a solução desse problema foram propostos e os resultados publicados na literatura científica. / Through color eye fundus images, the eye care specialists can detect possible complications related to diabetes as the vision impairment or vision loss. The Diabetic Macular Edema (DME) is the most common cause of vision damage in working-age people. Therefore, this thesis presents an approach to automatically identify the different levels of severity of diabetic macular edema aiming to assist the expert in the diagnosis of this pathology. As a final result, a methodology to automatically and quickly identify, from the eye fundus image, if a patient has the EMD mild, moderate or severe EMD is proposed. In a preliminary evaluation of our DME grading scheme using publicly available eye fundus images (i.e., DIARETDB1 image database), an accuracy of 94.29% was obtained. Some intermediate methods needed to solve this problem have been proposed and the results published in scientific literature.

This master´s thesis deals with blood vessel segmentation in retinal image data. The theoretical part is focused on the basic description of anatomy and physiology of the eye and methods of observing the back of the eye. This thesis also describes the principles of classical and convolutional neural networks and segmentation techniques that are used to segment blood vessel in retinal images. In the practical part, a segmentation method using convolutional neural network U-net is implemented. This neural network is trained on the three datasets. Two datasets include images from experimental video ophthalmoscope. Because it impossible to compare the results of these two datasets with any other methods of retinal blood vessel segmentation, U-net is trained on other dataset that is HRF database. This dataset includes fundus images. The results of testing on this dataset serves for comparing results with other methods of retinal blood vessel segmentation.

From biometrical point of view, main features of retina are fovea, optic nerve and blood vessel tree. Blood vessel tree is unique for each person and this biological feature is used in biometric systems for person-recognition by retinal images. This document describes methods for optic disc and fovea localization, method for vessel tree segmentation, which is based on well-known \emph{Matched filters} method and also describes method for localization of blood vessel bifurcations. Main goal of this thesis is creation of program which can automatically preprocess input image, segment blood vessels and localize vessel bifircations. The program is implemented in Java with OpenCV library.

No description available.