Cette thèse a pour objet l’étude de nouvelles méthodes de traitement d’image appliquées à l’analyse d’images numériques du fond d'œil de patients diabétiques. En particulier, nous nous sommes concentrés sur le développement algorithmique supportant un système de dépistage automatique de la rétinopathie diabétique. Les techniques présentées dans ce document peuvent être classées en trois catégories: (1) l’évaluation et l’amélioration de la qualité d’image, (2) la segmentation des lésions, et (3) le diagnostic. Pour la première catégorie, nous présentons un algorithme rapide permettant l’estimation numérique de la qualité d’une seule image à partir de caractéristiques extraites de la vascularisation et de la couleur du fond d'œil. De plus, nous démontrons qu’il est possible d’augmenter la qualité des images et de supprimer les artefacts de réflexion en fusionnant les informations extraites de plusieurs images d’un même fond d'œil (images capturées en changeant le point d’attention regardé par le patient). Pour la deuxième catégorie, deux familles de lésion sont ciblées: les exsudats et les microanévrysmes. Deux nouveaux algorithmes pour l’analyse des images du fond d'œil sont proposés et comparés avec les techniques existantes afin de démontrer leur efficacité. Dans le cas des microanévrismes, une nouvelle méthode basée sur la transformée de Radon a été développée. Dans la dernière catégorie, nous présentons un algorithme permettant de diagnostiquer la rétinopathie diabétique et les œdèmes maculaires en analysant les lésions détectées par segmentation d’image; à partir d’une seule image, notre algorithme permet de diagnostiquer une rétinopathie diabétique et/ou un œdème maculaire en ~ 22 secondes sur une machine à 1,6 GHz avec 4 Go de RAM; de plus, nous montrons les premiers résultats de notre algorithme de détection d'œdème maculaire basé sur des images du fond d'œil multiples, qui peut éventuellement permettre d’identifier le gonflement de la macula même si aucune lésion n’est visible. / In this Ph.D. thesis, we study new methods to analyse digital fundus images of diabetic patients. In particular, we concentrate on the development of the algorithmic components of an automatic screening system for diabetic retinopathy. The techniques developed can be categorized in: quality assessment and improvement, lesion segmentation and diagnosis. For the first category, we present a fast algorithm to numerically estimate the quality of a single image by employing vasculature and colour-based features; additionally, we show how it is possible to increase the image quality and remove reflection artefacts by merging information gathered in multiple fundus images (which are captured by changing the stare point of the patient). For the second category, two families of lesion are targeted: exudate and microaneurysms; two new algorithms which work on single fundus images are proposed and compared with existing techniques in order to prove their efficacy; in the microaneurysms case, a new Radon transform-based operator was developed. In the last diagnosis category, we have developed an algorithm that diagnoses diabetic retinopathy and diabetic macular edema based on the lesions segmented; starting from a single unseen image, our algorithm can generate a diabetic retinopathy and ma cular edema diagnosis in _22 seconds on a 1.6 GHz machine with 4 GB of RAM; additionally, we show the first results of a macular edema detection algorithm based on multiple fundus images, which can potentially identify the swelling of the macula even when no lesions are visible.

Orientador: Vital Paulino Costa / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Ciências Médicas / Made available in DSpace on 2018-08-24T05:19:20Z (GMT). No. of bitstreams: 1 Santana_AlanaMendonca_M.pdf: 2805390 bytes, checksum: 9512111e988de78a1ffde9d9f877658b (MD5) Previous issue date: 2014 / Resumo: O objetivo deste trabalho foi investigar os efeitos das combinações fixas dos análogos de prostaglandinas com timolol sobre a barreira hematoaquosa, a espessura macular central e a pressão intraocular (PIO), em pacientes pseudofácicos com glaucoma primário de ângulo aberto. Neste ensaio clínico randomizado, com observador mascarado e duração de 6 meses, os pacientes foram tratados uma vez por dia (20:00 horas) com lubrificante (grupo controle) ou com a combinação fixa de maleato de timolol 0,5% e latanoprosta 0,005% (CFLT), maleato de timolol 0,5% e bimatoprosta 0,03% (CFBT) ou maleato de timolol 0,5% e travoprosta 0,004% (CFTT). Foi incluído no estudo apenas um olho de 61 pacientes: CFLT (n=16), CFBT (n=15), CFTT (n=15) e grupo controle (n=15). A barreira hematoaquosa foi avaliada por meio do &quot;laser flare meter¿ antes do início do uso das medicações e após 15 dias, 1,2,3,4,5 e 6 meses de tratamento. A PIO foi medida sempre às 9:00 horas, nas mesmas ocasiões. A espessura macular central foi avaliada por meio da tomografia de coerência óptica antes do uso das medicações, após 1 e 6 meses de tratamento ou na hipótese de piora da acuidade visual. Não houve aumento estatisticamente significante nos valores médios de &quot;flare¿ em comparação aos valores iniciais em todos os grupos (p>0,05) em todas as visitas, exceto no grupo CFTT no primeiro mês (p=0,0476) e no grupo CFLT (p=0,0129) no terceiro mês de seguimento. Não houve diferença estatisticamente significativante entre os valores médios de &quot;flare¿ entre os grupos durante o estudo (p>0,05). A média dos valores da espessura macular central aumentou significativamente nos grupos CFLT (p=0,012) e CFTT (p=0,0419) no primeiro mês de tratamento em relação aos valores iniciais. Não houve aumento estatisticamente significante nos valores médios da espessura macular central em relação aos valores iniciais em todos os grupos após 6 meses de tratamento (p>0,05) e não ocorreu diferença estatisticamente significante entre os grupos durante o estudo (p>0,05). A PIO média inicial foi significativamente menor no grupo controle (p=0,0000). Todas as combinações fixas reduziram significativamente a PIO em todas as visitas (P<0,0001), com efeito hipotensor semelhante entre si (p=0,816). Estes resultados indicam que o uso das combinações fixas de análogos de prostaglandinas com timolol não aumentou significativamente a média dos valores de &quot;flare¿ ou a média da espessura macular após 6 meses de tratamento nesta população / Abstract: The aim of this study was to investigate the effects of prostaglandin analogues and timolol fixed-combinations on the blood-aqueous barrier, central macular thickness and intraocular pressure (IOP) in pseudophakic patients with primary open angle glaucoma. In this randomized, masked-observer, 6-month clinical trial, patients were treated once daily (8 pm) with lubricant eye drops (control group), 0,5% timolol and 0,005% latanoprost fixed-combination (LTFC), 0,5% timolol and 0,03% bimatoprost fixed-combination (BTFC) or 0,5% timolol and 0,004% travoprost fixed-combination (TTFC). One eye of 61 patients were included in the study: LTFC (n=16), BTFC (n=15), TTFC (n=15) and control group (n=15). The blood-aqueous barrier status was assessed using the laser flare meter before the medications were started and after 15 days, 1, 2, 3, 4, 5 and 6 months of treatment. The IOP was measured always at 9 am, at the same intervals. The central macular thickness was evaluated with optical coherence tomography before medications were started, after 1 and 6 months of follow-up or if a patient showed decreased visual acuity at any time during follow-up. There was no significant increase in mean flare measurements from baseline in all groups (p>0.05) in all visits, except TTFC group at 1 month (p=0,0476) and LTFC group at 3 months (p=0,0129). There were no significant differences in mean flare values among the groups (p>0.05). Mean central macular thickness values were significantly higher in LTFC (p=0,012) and TTFC groups (p=0,0419) at 1 month of follow-up. There was no significant increase in mean central macular thickness values from baseline in all groups (p>0.05) after 6 months and no significant differences among the groups (p>0.05) during follow-up. At baseline, mean IOP was significantly lower in control group (p=0,0000). All fixed-combinations significantly reduced IOP in all follow-up visits, with similar lowering effect (p=0,816). These findings indicate that the use of prostaglandin analogues and timolol fixed-combinations didn't significantly increased mean flare values or mean central macular thicknes measurements after 6 months of follow-up in this population / Mestrado / Oftalmologia / Mestra em Ciências Médicas

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22119号 / 医博第4532号 / 新制||医||1039(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 大森 孝一, 教授 Shohab YOUSSEFIAN, 教授 山下 潤 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第22320号 / 医博第4561号 / 新制||医||1041(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 椛島 健治, 教授 大森 孝一, 教授 森田 智視 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・論文博士 / 博士(医学) / 乙第12917号 / 論医博第2092号 / 新制||医||1009(附属図書館) / 32127 / 京都大学大学院医学研究科医学専攻 / (主査)教授 河野 憲二, 教授 大森 治紀, 教授 渡邉 大 / 学位規則第4条第2項該当 / Doctor of Medical Science / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(医学) / 甲第21629号 / 医博第4435号 / 新制||医||1033(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 伊佐 正, 教授 柳田 素子, 教授 宮本 享 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

Objetivo: Avaliação funcional (acuidade visual e sensibilidade ao contraste) e morfológica da retina de pacientes com edema macular diabético (EMD) tratados com bevacizumabe intravítreo e controle glicêmico optimizado durante 24 semanas de acompanhamento. Projeto: Estudo prospectivo intervencionista, randomizado e controlado. Métodos: Quarenta e um olhos de 34 pacientes com diabetes mellitus tipo 2 e EMD com Hemoglobina glicada (HbA1c) <= 11% receberam injeções de bevacizumabe intravítreo (BIV) com controle glicêmico optimizado nas semanas 0, 6, 12 e 18 (grupo 1; n=21); ou receberam simulação de BIV com controle glicêmico optimizado na semana 0 e 6 e, posteriormente, BIV com controle glicêmico optimizado na semana 12 e 18 (Grupo 2; n=20). Mudanças na sensibilidade ao contraste (SC), na acuidade visual (AV) e na espessura foveal (EF) medida por Tomografia de coerência óptica (TCO) foram comparadas antes do início do tratamento e nas semanas 2, 6, 12, 14, 18 e 24. A análise laboratorial do controle sistêmico foi comparada também no início e nas semanas 12 e 24. Resultados: O estudo mostrou AV e EF significativamente melhores em 24 semanas em ambos os grupos. Na 24ª semana, a média da SC melhorou no grupo 1 de 1,14±0,36logCS para 1,33±0,24logCS (P=0,002) e, também, no grupo 2 de 1,11±0,29 para 1,26±0,23logCS (P = 0,004) em 24 semanas, sem diferença significativa entre eles (P = 0,52). Houve uma redução média de 0,78% nos níveis de HbA1c (8,28±1,29% para 7,5±1,20%; P < 0,001) em 24 semanas. As maiores diferenças entre os grupos foram detectadas entre o pré-tratamento e a segunda semana, onde a EF diminuiu rapidamente no grupo 1 em comparação ao grupo 2 (116±115 vs. 17±71?m, respectivamente; P < 0,01). Em 12 semanas, as medidas já não foram estatisticamente significativas (variação da EF = grupo 1 91±184?m vs. grupo 2 62±152?m; P = 0.72). No pré-tratamento, houve correlação significativa entre SC e AV (r = -0,70; P < 0,001) e também entre SC e EF (r = -0,63; P < 0,001) que se manteve em 24 semanas. Conclusão: O uso do bevacizumabe intravítreo associado ao controle glicêmico optimizado melhorou significativamente a função visual, incluindo a sensibilidade ao contraste dos olhos com EMD. No entanto, o controle glicêmico optimizado isolado também se mostrou eficiente durante 12 semanas / Purpose: Functional (visual acuity and contrast sensitivity) and morphological evaluation of the retina of patients with diabetic macular edema (DME) treated with intravitreal bevacizumab and optimized glycemic control on 24-week followup. Project: Prospective, interventional, randomized controlled trial. Methods: Forty-one eyes of 34 patients with type 2 diabetes mellitus and DME with hemoglobin A1c (HbA1c) <= 11% received injections of intravitreal bevacizumab (IVB) with optimized glycemic control (group 1; n=21) at week 0, 6, 12 and 18; or received IVB simulation with optimized glycemic control at week 0 and 6 and, after three months, IVB with optimized glycemic control at week 12 and 18 (Group 2; n=20). Mean Change in contrast sensitivity (CS), best corrected visual acuity (BCVA) and central foveal thickness (CFT) - as measured by optical coherence tomography (OCT) were compared at baseline, and at weeks 2, 6, 12, 14, 18 and 24. A laboratory analysis of systemic control was also compared at baseline and at weeks 12 and 24. Results: The study showed that BCVA and CFT were significantly better in both groups at 24 weeks. Mean CS also improved significantly in group 1 from 1.14±0.36 to 1.33±0,24logCS (P=0,002) and also in group 2 from 1.11±0.29 to 1.26±0.23logCS (P=0,004) at 24 weeks, with no significant difference between them (P = 0.52). There was a mean reduction of 0.78% in HbA1c levels (8,28±1,29 to 7,5±1,20%; P < 0,001) at 24 weeks. The major differences between both groups were detected between the baseline and the second week, whem CFT promptly decreased in group 1 compared to group 2 (116±115 ?m vs. 17±71 ?m; P < 0,01). At 12 weeks, the measures were no longer statistically significant (variation of CFT = group 1 91±184 ?m vs. group 62±152 ?m, P = 0.72). At baseline, there was meaningful correlation between CS and BCVA (r = -0,70; P < 0,001) and also between CS and CFT (r = -0.63; P < 0.001) that remained at 24 weeks. Conclusion: The use of intravitreal bevacizumabe associated with optimized glycemic control improved significantly the visual function, including contrast sensitivity in eyes with DME. However, optimized glycemic control alone also shown to be efficient at 12 weeks

No description available.

In this Ph.D. thesis, we study new methods to analyse digital fundus images of diabetic patients. In particular, we concentrate on the development of the algorithmic components of an automatic screening system for diabetic retinopathy. The techniques developed can be categorized in: quality assessment and improvement, lesion segmentation and diagnosis. For the first category, we present a fast algorithm to numerically estimate the quality of a single image by employing vasculature and colour-based features; additionally, we show how it is possible to increase the image quality and remove reflection artefacts by merging information gathered in multiple fundus images (which are captured by changing the stare point of the patient). For the second category, two families of lesion are targeted: exudate and microaneurysms; two new algorithms which work on single fundus images are proposed and compared with existing techniques in order to prove their efficacy; in the microaneurysms case, a new Radon transform-based operator was developed. In the last diagnosis category, we have developed an algorithm that diagnoses diabetic retinopathy and diabetic macular edema based on the lesions segmented; starting from a single unseen image, our algorithm can generate a diabetic retinopathy and ma cular edema diagnosis in _22 seconds on a 1.6 GHz machine with 4 GB of RAM; additionally, we show the first results of a macular edema detection algorithm based on multiple fundus images, which can potentially identify the swelling of the macula even when no lesions are visible.

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.