China
Africa
Explore chapters and articles related to this topic
Dual Customized U-Net-based Automated Diagnosis of Glaucoma
Published in K. Gayathri Devi, Kishore Balasubramanian, Le Anh Ngoc, Machine Learning and Deep Learning Techniques for Medical Science, 2022
C. Thirumarai Selvi, J. Amudha, R. Sudhakar
Artem Sevastopolsky [21] has developed alterations to the U-Net convolutional neural method for detecting glaucoma using the optic cup and disc segmentation. This deep learning methodology has experimented on DRISHTI-GS, DRIONS-DN, RIM ONE v.3 publically available data sets. This technique outperforms the other state of art techniques by prediction time. Maninis et al [22] have demonstrated deep-learning-based optic disc segmentation of retinal images. This technology applies a transfer learning procedure and VGG16 based fully convolutional neural network. They achieved human expert quality of dice score and boundary error. Zilly et al [23] have recorded the fundus image optic-cup and disc segmentation by employing improvement in the convolutional filter with entropy sampling. Adaboost classifier has used in the hierarchical network for mosaic slab classification was found by Dogan and Akay [24]. Maninis [22] method has the drawback of longer execution time for training, large model size, and requires more General Processing Unit (GPU) memory. Zilly [23] techniques are more complicated, difficult to program and to produce results. In the entropy sampling method, images have to be cropped for the optic disc area before the training. Artem's [21] work has focused on lesser prediction time, but the edges are more delicate to bring more accurate classification. It further requires some enhancement in optic cup segmentation.
Results and Discussion
Published in Arwa Ahmed Gasm Elseid, Alnazier Osman Mohammed Hamza, Computer-Aided Glaucoma Diagnosis System, 2020
Arwa Ahmed Gasm Elseid, Alnazier Osman Mohammed Hamza
The optic disc is the pallor circular region located at the position where the optic nerve leaves the eye, where the optic cup is the central, bright, yellowish circular region in the optic disc. Optic disc segmentation: the proposed method can be divided into 3 steps. The first step is to remove the vessel to achieve accurate segmentation, and it is done by morphological operation. The second step is thresholding and binarization, and the given image will convert to a binary image. From this image, the optic disc boundary can be easily extracted.
The presentation and management of physical disease in older people
Published in David Beales, Michael Denham, Alistair Tulloch, Community Care of Older People, 2018
Occasionally visual problems can present as a true or apparent confusional state. The basis of this is either the confusional effect of cortical blindness (or a visual agnosia) so that the subject is unaware that they are not seeing properly, or as a result of a painful eye condition. Acute glaucoma normally presents as an acute painful eye, but in patients with cognitive impairment any painful condition may not be accurately reported and may present as worsening confusion. Part of the normal examination of the older adult should include a simple test of visual acuity (such as reading a newspaper), assessment of the visual fields to confrontation and fundoscopy to assess the area of the optic cup (as a screening test for glaucoma). About 10% of newly diagnosed elderly diabetics will have detectable retinopathy.
Optical coherence tomography in the assessment of simultaneous macula oedema and papilloedema
Published in Clinical and Experimental Optometry, 2020
Amanda Edgar, Jayson Ward, Craig Woods
The spectrum of causes of papilloedema entail a variation of non‐specific presenting symptoms. These include nausea, headaches, vomiting, visual disturbances and behavioural changes, although cases of papilloedema may also present asymptomatically.2001 Evidence for diagnosis is strongly supported by a posterior ocular examination finding of bilateral blurred ONH margins, elevation, filling in of the optic cup and the absence of spontaneous venous pulsations at the emergence of veins from the optic cup.2001 This visualisation can be graded subjectively by the examining practitioner using grading scales; however, many of these scales are limited by the need for validation.1982 Another tool that may offer advantage to the practitioner is the use of the quantitative analysis of ocular structures performed by optical coherence tomography (OCT).1982
Machine Learning in the Detection of the Glaucomatous Disc and Visual Field
Published in Seminars in Ophthalmology, 2019
David J. Smits, Tobias Elze, Haobing Wang, Louis R. Pasquale
Automated glaucomatous nerve detection from fundus photographs has remained a difficult task due to a number of factors. Color, focus, magnification, photo quality and, most importantly, natural variation of optic nerves including the spectrum of normal CDR in a population and between populations39 all contribute to the challenge. Previous attempts at automatic detection utilized a process called image auto-segmentation. Auto-segmentation attempts to partition an image into segments that can be meaningful. It does so by clustering pixels with similar attributes such as color or intensity. This can serve to define boundaries, such as the anatomic border between the optic disc and optic cup. Researchers have used auto-segmentation in an attempt to determine CDR40;−43 however, peripapillary atrophy and vasculature patterns of the normal optic nerve head can present significant challenges.
Automatic Assessment of Biometric Parameters in Optic Nerve Head Area by “Zhongshan ONH Calculator (ZOC)”
Published in Current Eye Research, 2019
Fei Li, Kai Yu, Lichun Zhang, Kai Gao, Xinjian Chen, Xiulan Zhang
Glaucoma leads to irreversible blindness through direct damage to the optic nerve.1 Ocular hypertension is one of the main risk factors of glaucoma, which affects blood flow in the optic nerve head (ONH) area and triggers apoptosis of ganglion cells.2 In the early stage of glaucoma, thinning of the retinal nerve fiber layer (RNFL) may be detected with an optical coherence tomography (OCT) scan of the ONH area.3,4 With the progression of glaucoma, the RNFL will gradually become thinner, which is a useful landmark in the diagnosis and prediction of glaucoma. In addition to the RNFL, many other morphological changes are found at the posterior segment of human eyes, such as the optic cup, lamina cribrosa (LC), rim area (RA), etc. The LC is where the optic nerve penetrates the eyeball, which also bears abnormally high intraocular pressure (IOP) in glaucoma patients. As a result, distinct thinning of the LC can also be observed in all types of glaucoma5; the optic cup refers to the small concavity in the center of the optic disk, whose depth and volume are affected by glaucoma progression.