|
|
Detection of pathologies in retinal images
Mesíková, Klaudia ; Kolář, Radim (referee) ; Vičar, Tomáš (advisor)
The goal of this thesis is to design and implement software for the detection of diabetes mellitus symptoms from the image of the human eye retina. Diabetic retinopathy is the most common disease affecting the retina. Pathologies connected with this disease can lead to partial or complete blindness. For the detection of pathological symptoms is important to correctly detect some parts of the eye retina such as optic disc and blood vessels. These can cause a problem with the identification of disease. After removing the optic disc and blood vessels, the pathology object is being detected.
|
|
|
Illness Detection in Eye Retina Image
Aubrecht, Tomáš ; Drahanský, Martin (referee) ; Semerád, Lukáš (advisor)
Age-related macular degeneration is one of the most common retinal diseases of the human eye that leads at different rates to blindness. This disease occurs in two forms. While the wet can slow down its progression, for dry form there is no available treatment method so far, so prevention is the most important. For this reason, the aim of this work is to design and implement software that allows automatic detection of the symptoms of this disease from retinal images. The symptom detection algorithm is based on adaptive thresholding which obtains suspicious areas that are subsequently categorized. 407 retinal images were used for the evaluation and the software was highly successful. When used in conjunction with an optical retinal scanner, it helps ophthalmologists, for example, to diagnose more quickly.
|
|
|
Automatic Detection of Eye Retinal Pathologies
Tlustoš, Vít ; Beran, Vítězslav (referee) ; Drahanský, Martin (advisor)
Cílem této práce je navrhnout a vyvinout systém, který automaticky odhalí vybrané patologie nacházející se na snímcích sítnice lidského oka. Sítnice jako jediný orgán v těle obsahuje světlocitlivé buňky potřebné k vidění. Proto, aby byla léčba onemocnění sítnice úspěšná je klíčové jeho včasné zachycení a přesné určení rozsahu. Navržený systém automaticky k dodanému snímku vygeneruje segmentační masky reprezentující výskyt jednotlivých patologií. Výsledek je poté prezentován uživateli. O samotné vyhodnocení se stará konvoluční neuronová síť jejímž základem je architektura U-Net. Síť byla natrénovaná na datasetu Indian Diabetic Retinopathy Image Dataset zkráceně IDRiD, který obsahuje 81 snímků sítnice a k nim příslušících anotací. Úspěšnost navrhovaného systému byla stanovena pomocí AUC-PR skóre (plocha pod precision-recall křivkou). Segmentace tvrdých exsudátů, měkkých exsudátů, hemoragií a mikroaneuryzmat dosáhla hodnot AUC-PR 74%, 50%, 45% a 33%, v daném pořadí. Tato práce přináší inovativní architekturu, která má v případě dalšího rozvoje potenciál být využita oftalmology pro diagnostiku a stanovení rozsahu onemocnění sítnice oka.
|
|
|
Disease Detection in Eye Retina Image
Koštialik, Daniel ; Semerád, Lukáš (referee) ; Maruniak, Lukáš (advisor)
Diabetic retinopathy and age related macular degeneration are among the most common eye retina diseases, which cause partial or complete blindness. The main goal of this thesis is to design and implement software for automatic detection of symptoms from eye fundus images. The detection algorithm is based on the image segmentation by region growing method and afterwards analysis. Determination of retina objects such as optic disc, macula and blood vessels is important prior symptoms detection as they can adversely affect the results of the analysis. Total 259 images were analysed and algorithm reaches more than 90 % average success rate. The algorithm, in combination with appropriate hardware and optic mechanism, forms one of practical application in global population screening. Thanks the automatic detection it is possible to determine the presence of symptoms and start an early treatment.
|
|
|
Detection of Diseases of Diabetes on the Human Eye Retina
Sýkorová, Tereza ; Semerád, Lukáš (referee) ; Drahanský, Martin (advisor)
This bachelor's thesis deals with the detection of the symptoms of diabetic retinopathy at retinal images taken by a digital fundus camera. Optic disc, fovea, and blood vessels are found before searching for exudates and hemorrhages. This step improves final detection. The detector uses morphological reconstruction of a candidate region for determination of specific lesions. An algorithm based on thresholding precises its edges. Found regions are classified according to shape and color. Evaluation of detection was done using 120 images selected from three databases. Adding automatic detection of signs of diabetic retinopathy into equipment for retinal screening can help medical doctors in diagnosis and prevent possible vision loss which the disease can cause.
|
|
|
Detection and Recognition of Diabetes Disease Impacts to the Human Eye Retina
Jausch, Andrej ; Semerád, Lukáš (referee) ; Drahanský, Martin (advisor)
This bachelor's thesis deals with the design of algorithms for the recognition of a diabetes disease impacts to the human eye retina. Diabetic retinopathy is one of the most common diseases aecting the retina and its consequences lead to partial or complete weakness. The basis of the algorithm for detection is to create candidate areas from dierent viewpoints of image processing - computer vision and their subsequent analysis. Core components of the retina have impacts to detection results - optical disc and blood vessels, which need to be properly detected and subsequently excluded from processing. Testing the implemented application took place in 68 images selected from two databases. One of the possible uses of the proposed methods in the future is in combination with the retinal scanning device for the automatic detection of diabetes symptoms during the retinal screening process.
|
|
|
Comparison of Retinal Images with Pathological Findings
Palacková, Bianca ; Semerád, Lukáš (referee) ; Drahanský, Martin (advisor)
The goal of this thesis is to design and implement software for comparison of retinal images with pathological findings. The most common diseases affecting the retina are diabetic retinopathy and age related macular degeneration. Detection of main components such as optic disc and fovea, need to be detected for proper comparison and detection of diseases. 570 images was used for evaluation of detection of these main components. In both cases, algorithm achieved success over 90\%. 120 images were analysed by 3 ophthalmologists for evaluation of ability to locate pathological findings. Automatic comparison of retinal images can be useful for determination of disease progression.
|
|
|
Reconstruction of Extracted Bloodstream in Images of Retinas
Kozel, Vojtěch ; Drahanský, Martin (referee) ; Semerád, Lukáš (advisor)
Retinal bloodstream plays a significant role in many specializations. In medicine, retinal images are used for automatic disease diagnosis. The blood vessel tree is unique for each individual, and as such this feature is often used in biometric systems for person-recognition. Healthy individuals possess consistent retinas throughout their life, however, there are many reasons why retinal changes may occur. The most common reason for physical changes is disease. In such cases problems arise in automated processing of retina images. These problems may also lie with retinal scans errors or blood vessel extraction algorithm error. This thesis describes reasons why segmented blood vessels are interrupted. Main goal of this thesis is to create a program which can automatically locate interrupted vessel segments and reconstruct them. The program is implemented in Java with OpenCV library.
|
|
|
Generation of Synthetic Retinal Images with High Resolution
Aubrecht, Tomáš ; Heidari, Mona (referee) ; Drahanský, Martin (advisor)
K pořízení snímků sítnice, která představuje nejdůležitější část lidského oka, je potřeba speciálního vybavení, kterým je fundus kamera. Z tohoto důvodu je cílem této práce navrhnout a implementovat systém, který bude schopný generovat takovéto snímky bez použítí této kamery. Navržený systém využívá mapování vstupního černobílého snímku krevního řečiště sítnice na barevný výstupní snímek celé sítnice. Systém se skládá ze dvou neuronových sítí: generátoru, který generuje snímky sítnic, a diskriminátoru, který klasifikuje dané snímky jako reálné či syntetické. Tento systém byl natrénován na 141 snímcích z veřejně dostupných databází. Následně byla vytvořena nová databáze obsahující více než 2,800 snímků zdravých sítnic v rozlišení 1024x1024. Tato databáze může být použita jako učební pomůcka pro oční lékaře nebo může poskytovat základ pro vývoj různých aplikací pracujících se sítnicemi.
|
|
|
Tool for Detection and Correction of Images with Diseased Eye Retinas
Jochlík, Jakub ; Semerád, Lukáš (referee) ; Drahanský, Martin (advisor)
Loss or partial loss of eye sight can have major effect on quality of person's life. One of the most common diseases, which causes loss or partial loss of eye sight are diabetic retinopathy and age releated macular degeneration. Both of these diseases can be prevented or mediated by early detection and proper treatment. The fundus camera, which is used to capture eye retina, has had major effect on increasing quality and speed of early detection. Images captured by fundus camera can be automatically analyzed in order to detect any possible signs of retina damage. This thesis proposes one possible way of automating this process. First part of this thesis describes eye, its diseases and capturing technology. Second part then proposes way of automating detection process and its implementation. Lastly, the results are evaluated.
|
guest ::
