- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
[IEEE 2017 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) - Atlanta, GA (2017.10.21-2017.10.28)] 2017 IEEE Nuclear Science Symposium and Medical Imaging Conference (NSS/MIC) - Automated Optic Nerve Head Detection Based on Different Retinal Vasculature Segmentation Methods and Mathematical Morphology
摘要： Computer vision and image processing techniques provide important assistance to physicians and relieve their work load in different tasks. In particular, identifying objects of interest such as lesions and anatomical structures from the image is a challenging and iterative process that can be done by using computer vision and image processing approaches in a successful manner. Optic Nerve Head (ONH) detection is a crucial step in retinal image analysis algorithms. The goal of ONH detection is to ?nd and detect other retinal landmarks and lesions and their corresponding diameters, to use as a length reference to measure objects in the retina. The objective of this study is to apply three retinal vessel segmentation methods, Laplacian-of-Gaussian edge detector, Canny edge detector, and Matched ?lter edge detector for detection of the ONH either in the normal fundus images or in the presence of retinal lesions (e.g. diabetic retinopathy). The steps for the segmentation are as following: 1) Smoothing: suppress as much noise as possible, without destroying the true edges, 2) Enhancement: apply a ?lter to enhance the quality of the edges in the image (sharpening), 3) Detection: determine which edge pixels should be discarded as noise and which should be retained by thresholding the edge strength and edge size, 4) Localization: determine the exact location of an edge by edge thinning or linking. To evaluate the accuracy of our proposed method, we compare the output of our proposed method with the ground truth data that collected by ophthalmologists on retinal images belonging to a test set of 120 images. As shown in the results section, by using the Laplacian-of-Gaussian vessel segmentation, our automated algorithm ?nds 18 ONHs in true location for 20 color images in the CHASE-DB database and all images in the DRIVE database. For the Canny vessel segmentation, our automated algorithm ?nds 16 ONHs in true location for 20 images in the CHASE-DB database and 32 out of 40 images in the DRIVE database. And lastly, using matched ?lter in the vessel segmentation, our algorithm ?nds 19 ONHs in true location for 20 images in CHASE-DB database and all images in the DRIVE.
关键词： Laplacian-of-Gaussian edge detector,Diabetic retinopathy,Match ?lter,image processing,Optic Nerve Head,Canny edge detector,retinal blood vessel
更新于2025-09-09 09:28:46
-
[IEEE 2018 24th International Conference on Pattern Recognition (ICPR) - Beijing, China (2018.8.20-2018.8.24)] 2018 24th International Conference on Pattern Recognition (ICPR) - Early Diagnosis of Diabetic Retinopathy in OCTA Images Based on Local Analysis of Retinal Blood Vessels and Foveal Avascular Zone
摘要： This paper introduces a diagnosis system for detecting early signs of diabetic retinopathy (DR) using optical coherence tomography angiography (OCTA) images. We developed a segmentation technique that was able to extract blood vessels from both retinal superficial and deep maps. It is based on a higher order joint Markov-Gibbs random field (MGRF) model, which combines both current and spatial appearance information of retinal blood vessels. To be able to train/test a support vector machine (SVM) classifier, three local features were extracted from the segmented images. These extracted features are the density and appearance of the retinal blood vessels in addition to the distance map of the foveal avascular zone (FAZ). Then, we used SVM with linear kernel to distinguish sub-clinical DR patients from normal cases. By using 105 subjects, the presented computer-aided diagnosis (CAD) system demonstrated an overall accuracy (ACC) of 97.3% and a Dice similarity coefficient (DSC) of 97.9%.
关键词： Markov-Gibbs random field,support vector machine,diabetic retinopathy,optical coherence tomography angiography,computer-aided diagnosis
更新于2025-09-09 09:28:46
-
Retinal Neurodegeneration in Diabetic Patients Without Diabetic Retinopathy
摘要： PURPOSE. To compare the thickness of all retinal layers between a nondiabetic group and diabetic patients without diabetic retinopathy (DR). METHODS. Cross-sectional study, in which all subjects underwent an ophthalmic examination including optical coherence tomography. After automatic retinal segmentation, each retinal layer thickness (eight separate layers and overall thickness) was calculated in all nine Early Treatment Diabetic Retinopathy Study (ETDRS) areas. The choroidal thickness (CT) also was measured at five locations. Generalized additive regression models were used to analyze the data. RESULTS. A total of 175 patients were recruited, 50 nondiabetic subjects and 125 diabetic patients without DR, stratified into three groups according to diabetes duration: group I (<5 years, n = 55), group II (5–10 years, n = 39), and group III (>10 years, n = 31). Overall, groups I and III of diabetic patients had a decrease in the photoreceptor layer (PR) thickness, when compared with the nondiabetic subjects in six ETDRS areas (P < 0.0007). Patients with more recent diagnosis (group I) had thinner PR than those with moderate duration (group II). Interestingly, patients with longer known disease (group III) had the thinnest PR values. There were no overall differences in the remaining retinal parameters. CONCLUSIONS. Retinal thickness profile is not linear throughout disease duration. Even in the absence of funduscopic disease, PR layer in diabetic patients seems to differ from nondiabetic subjects, thus suggesting that some form of neurodegeneration may take place before clinical signs of vascular problems arise.
关键词： neurodegeneration,retinal layers,diabetic retinopathy,optical coherence tomography
更新于2025-09-09 09:28:46
-
Expanding the role of medical retina virtual clinics using multimodal ultra-widefield and optical coherence tomography imaging
摘要： Background: An increasingly elderly population with a corresponding increase in ophthalmic conditions has led to increased pressure on hospital eye services (HES). In this study, we evaluated the use of a medical retina virtual clinic (MRVC), which has expanded into assessing all new medical retina referrals, where the need for urgent treatment was not clear. Methods: Retrospective analysis of all new patients who were seen in the MRVC between April 2016 and May 2018. Pro forma sheets were used in the MRVC to record the patient history, visual acuity, and type of imaging required. Two consultants reviewed the completed pro formas and images and provided a final diagnosis and management plan. These results and reasons for face-to-face (F2F) clinic appointment requests were analyzed. Results: Six hundred ten new referrals were enrolled in the virtual clinic. The most common diagnosis was diabetic eye disease (59.9%). In the virtual clinic 44.1% were followed up, 28.1% were discharged, and 27.8% were booked an F2F clinic appointment (urgent/routine). The main reason for F2F clinic was to offer treatment. Urgent F2F appointments took place on average 11.9 days after virtual clinic attendance. In only two cases was the image quality felt to be inadequate to assess the retina. Conclusions: MRVC is an effective way of triaging medical retina referrals to allow those patients needing treatment to be seen promptly in the medical retinal service. The use of multimodal ultra-widefield and optical coherence tomography imaging allows assessment of a wide range of retinal pathologies and is a promising solution to alleviate the burden on HES.
关键词： teleophthalmology,diabetic retinopathy,optical coherence tomography,macular degeneration,choroidal naevus,virtual clinic
更新于2025-09-09 09:28:46
-
Retrobulbar blood flow changes in eyes with diabetic retinopathy a 10 year follow up study
摘要： Purpose: We sought to assess long-term changes in the flow parameters of retrobulbar vessels in diabetic patients. Methods: The retrobulbar circulation of 138 eyes was evaluated between 1994 and 1995 and 36 eyes were reevaluated between 2004 and 2008 (study group). They were divided into four groups: eyes of diabetic patients without diabetic retinopathy (DR), eyes with nonproliferative DR, eyes with proliferative DR, and eyes of nondiabetic patients (controls). Color Doppler imaging was used to assess the flow velocities in the major retrobulbar vessels. The resistive index (RI) was calculated and compared among the groups and between the two time periods. Results: RI values of the central retinal artery and posterior ciliary artery had increased in the two non-DR groups and in the nonproliferative DR group, with a surprising decrease measured in eyes with proliferative DR (P= nonsignificant [NS]). Combining the nonproliferative DR and proliferative DR groups resulted in a milder increase of the RI of the posterior ciliary artery (P= NS) and the central retinal artery (P=0.02) in the DR group compared to the other groups. Conclusion: Our results demonstrate that an increase of the resistance in the retrobulbar vessels, as a part of DR, can lessen over time and may even be reversed.
关键词： color Doppler imaging,retrobulbar blood flow,diabetic retinopathy
更新于2025-09-09 09:28:46
-
Clinical Analysis of Newly Diagnosed Diabetes Mellitus Patients by Abnormal Fundus Examination
摘要： Purpose: To investigate the clinical analysis of newly diagnosed diabetes mellitus (NDM) patients with abnormal fundus examination at the first visit. Methods: This retrospective study utilized the first visit medical records of 15 patients (30 eyes) who were diagnosed with NDM from February 2011 to October 2016. Results: Patients were divided into 3 groups: 1) diabetic retinopathy group including proliferative diabetic retinopathy (PDR) (3) and severe non-proliferative diabetic retinopathy (NPDR) (1); 2) retinal vascular disease group including central retinal vein occlusion (CRVO) (1), branch retinal vein occlusion (1), vitreous hemorrhage with CRVO (1) and macular edema (1); and 3) other retinal disease group including vitreous hemorrhage due to choroidal neovascular rupture (1), exudative age-related macular degeneration (3), central serous chorioretinopathy (2), and macular hole (1). All 3 PDR patients had latent autoimmune diabetes in adults (type 1.5 diabetes). The remaining 12 patients had type 2 diabetes. Three patients showed mild NPDR in the opposite eye and the other 9 patients did not have diabetic retinopathy in the opposite eye. Onset age, HbA1C and proteinuria were significantly different between the diabetic retinopathy group and the other retinal disease group (p = 0.006, p = 0.012 and p = 0.006, Mann-Whitney test). Conclusions: In patients with various retinal diseases, early detection of NDM could be achieved by performing fundoscopic imaging and systemic examination as well as basic ophthalmologic examination. In addition, patients with diabetic retinopathy should be treated promptly through ophthalmology and internal medicine consultation. For the retinal vascular disease and other retinal disease groups, not only treatment for ophthalmic diseases, but also education about diabetes treatment are important.
关键词： Latent autoimmune diabetes in adults,Newly diagnosed diabetes mellitus,Diabetic retinopathy,HbA1C
更新于2025-09-09 09:28:46
-
Exporting Diabetic Retinopathy Images from VA VistA Imaging for Research
摘要： The US Department of Veterans Affairs has been acquiring store and forward digital diabetic retinopathy surveillance retinal fundus images for remote reading since 2007. There are 900+ retinal cameras at 756 acquisition sites. These images are manually read remotely at 134 sites. A total of 2.1 million studies have been performed in the teleretinal imaging program. The human workload for reading images is rapidly growing. It would be ideal to develop an automated computer algorithm that detects multiple eye diseases as this would help standardize interpretations and improve efficiency of the image readers. Deep learning algorithms for detection of diabetic retinopathy in retinal fundus photographs have been developed and there are needs for additional image data to validate this work. To further this research, the Atlanta VA Health Care System (VAHCS) has extracted 112,000 DICOM diabetic retinopathy surveillance images (13,000 studies) that can be subsequently used for the validation of automated algorithms. An extensive amount of associated clinical information was added to the DICOM header of each exported image to facilitate correlation of the image with the patient’s medical condition. The clinical information was saved as a JSON object and stored in a single Unlimited Text (VR = UT) DICOM data element. This paper describes the methodology used for this project and the results of applying this methodology.
关键词： VistA,Image retrieval for research,DICOM,Retinal imaging,JSON,Diabetic retinopathy
更新于2025-09-09 09:28:46
-
[IEEE 2018 International Conference on Soft-computing and Network Security (ICSNS) - Coimbatore, India (2018.2.14-2018.2.16)] 2018 International Conference on Soft-computing and Network Security (ICSNS) - Severity level detection of diabetic retinopathy using ELM classifier
摘要： An eye disease which destroys the normal vision ability of diabetic patients is known as diabetic retinopathy. Early diagnosis of this disease is necessary because, it is severe in the later stages. The presence exudates, micro aneurysms(MAs) and hemorrhages are the first clinical symptoms of this disease. Exudates are red dots formed by swelling of the weak part of the capillary wall. The detection of exudate in retinal fundus images and the severity level is an important task in applications such as diabetic retinopathy screening and early treatment. Diabetic retinopathy is identified by pouring chemical solution to the eye and then capturing the dilated image of the patient’s eye. This process causes irritation to the patients. This paper proposes a method to find the severity level of diabetic retinopathy. It uses non-dilated retinal fundus image to help ophthalmologists diagnose the disease. The exudates from the low contrast images are detected. A neighborhood based segmentation technique is used for localizing the exudates from the images. A support vector machine (SVM) and Extreme learning Machine (ELM) are used as the classifiers. The method assess the severity of the disease. The average classification accuracy for the ELM is 94.76%.
关键词： Color fundus images,ELM classifier,Micro aneurysms,Diabetic retinopathy
更新于2025-09-09 09:28:46
-
Multimodal imaging of foveal neovascularisation in diabetic retinopathy
摘要： Proliferative diabetic retinopathy is characterised by the growth of neovascular tissue, which almost always occurs at the optic disc and/or near the major nasal and temporal vascular arcades, sparing the foveal area.1 Finkelstein et al. ?rst described neovascular frond at the fovea arising from the perifoveal capillaries.2 A few other authors like Joondeph et al., Kurz et al., Rajagopal et al. and Seth et al. have also reported similar ?ndings.3–6 We report a rare case of unilateral foveal neovascularisation (NVF) in a middle-aged (40 years) woman of Asian origin who presented to us with chief complaints of blurring of vision in both eyes (left more than right) over the past few weeks, which was sudden in onset and progressive in nature. The patient had type 2 diabetes mellitus and dyslipidaemia for the past nine years. Her HbA1c was 10.5 per cent. On examination, her best-corrected visual acuity was 6/12 in the right eye and 6/24 in the left eye. The anterior segment was within normal limits for her age. Meticulous examination of the iris did not reveal any sign of neovascularisation. Posterior segment examination revealed proliferative diabetic retinopathy changes in both eyes with vitreous haemorrhage in the left eye only. The right eye was noted to have a whitish tuft at the fovea on biomicroscopy which turned out to be vascular in nature with distorted foveal avascular zone (FAZ) in both multicolor (Figure 1) and re?ectance (Figure 2) image. This was con?rmed with fundus ?uorescein angiography (confocal scanning laser ophthalmoscope-based; Figure 3) and spectral domain optical coherence tomography (Figure 4). Unlike previous reports, except for the recent report by Seth et al.,6 this patient had non-insulin-dependent diabetes mellitus. It is possible that this type of diabetes mellitus is not a predictor of NVF, but rather compromised ocular blood ?ow and duration and control of diabetes are predictors as reported by Summanen et al.7 In concordance with Finkelstein et al.,2 Joondeph et al.3 and Seth et al.,6 we noted macular ischaemia in our case, whereas Kurz et al.4 and Rajagopal et al.5 reported cases with perfused macula. Our case had unilateral NVF as reported by Joondeph et al.,3 Kurz et al.4 and Seth et al.,6 whereas Finkelstein et al.2 and Kurz et al.4 reported bilateral cases. We report for the ?rst time multimodal imaging of foveal neovascularisation in diabetic retinopathy. Limitations include a lack of early-phase fundus ?uorescein angiography imaging, green re?ectance, and optical coherence tomography angiography images. The early phase of ?uorescein angiography would have helped us with more effective visualisation of distorted FAZ and leakage from the NVF. Green re?ectance utilises a longer wavelength as compared to blue and has less absorption by the crystalline lens and xanthophyll pigments (at fovea) and therefore better delineates the fovea. In our case, this would have been the better imaging modality than blue re?ectance. Optical coherence tomography angiography beyond being non-invasive would have helped us to portray the FAZ and origin of foveal neovascularisation in a more desirable way, which we lacked.
关键词： multimodal imaging,diabetic retinopathy,foveal neovascularisation
更新于2025-09-09 09:28:46
-
The Effects of Diabetic Retinopathy and Pan-Retinal Photocoagulation on Photoreceptor Cell Function as Assessed by Dark Adaptometry
摘要： The pathophysiology of vision loss in persons with diabetic retinopathy (DR) is complex and incompletely defined. We hypothesized that retinal pigment epithelium (RPE) and rod and cone photoreceptor dysfunction, as measured by dark adaptometry, would increase with severity of DR, and that pan-retinal photocoagulation (PRP) would exacerbate this dysfunction. Dark adaptation (DA) was measured in subjects with diabetes mellitus and healthy controls. Dark adaptation was measured at 58 superior to the fovea following a flash bleach, and the data were analyzed to yield cone and rod sensitivity curves. Retinal layer thicknesses were quantified using spectral-domain optical coherence tomography (OCT). The sample consisted of 23 controls and 73 diabetic subjects. Subjects with moderate nonproliferative diabetic retinopathy (NPDR) exhibited significant impairment of rod recovery rate compared with control subjects (P ? 0.04). Cone sensitivity was impaired in subjects with proliferative diabetic retinopathy (PDR) (type 1 diabetes mellitus [T1DM]: P ? 0.0047; type 2 diabetes mellitus [T2DM]: P < 0.001). Subjects with untreated PDR compared with subjects treated with PRP exhibited similar rod recovery rates and cone sensitivities. Thinner RPE as assessed by OCT was associated with slower rod recovery and lower cone sensitivity, and thinner photoreceptor inner segment/outer segment layer was associated with lower cone sensitivity. The results suggest that RPE and photoreceptor cell dysfunction, as assessed by cone sensitivity level and rod- and RPE-mediated dark adaptation, progresses with worsening DR, and rod recovery dysfunction occurs earlier than cone dysfunction. Function was preserved following PRP. The findings suggest multiple defects in retinoid function and provide potential points to improve visual function in persons with PDR.
关键词： pan-retinal photocoagulation,dark adaptation,diabetic retinopathy
更新于2025-09-09 09:28:46