-
BMC Medical Imaging Jan 2021Glaucoma is an eye disease that causes vision loss and even blindness. The cup to disc ratio (CDR) is an important indicator for glaucoma screening and diagnosis....
BACKGROUND
Glaucoma is an eye disease that causes vision loss and even blindness. The cup to disc ratio (CDR) is an important indicator for glaucoma screening and diagnosis. Accurate segmentation for the optic disc and cup helps obtain CDR. Although many deep learning-based methods have been proposed to segment the disc and cup for fundus image, achieving highly accurate segmentation performance is still a great challenge due to the heavy overlap between the optic disc and cup.
METHODS
In this paper, we propose a two-stage method where the optic disc is firstly located and then the optic disc and cup are segmented jointly according to the interesting areas. Also, we consider the joint optic disc and cup segmentation task as a multi-category semantic segmentation task for which a deep learning-based model named DDSC-Net (densely connected depthwise separable convolution network) is proposed. Specifically, we employ depthwise separable convolutional layer and image pyramid input to form a deeper and wider network to improve segmentation performance. Finally, we evaluate our method on two publicly available datasets, Drishti-GS and REFUGE dataset.
RESULTS
The experiment results show that the proposed method outperforms state-of-the-art methods, such as pOSAL, GL-Net, M-Net and Stack-U-Net in terms of disc coefficients, with the scores of 0.9780 (optic disc) and 0.9123 (optic cup) on the DRISHTI-GS dataset, and the scores of 0.9601 (optic disc) and 0.8903 (optic cup) on the REFUGE dataset. Particularly, in the more challenging optic cup segmentation task, our method outperforms GL-Net by 0.7[Formula: see text] in terms of disc coefficients on the Drishti-GS dataset and outperforms pOSAL by 0.79[Formula: see text] on the REFUGE dataset, respectively.
CONCLUSIONS
The promising segmentation performances reveal that our method has the potential in assisting the screening and diagnosis of glaucoma.
Topics: Deep Learning; Diagnostic Techniques, Ophthalmological; Glaucoma; Humans; Image Processing, Computer-Assisted; Optic Disk
PubMed: 33509106
DOI: 10.1186/s12880-020-00528-6 -
Romanian Journal of Ophthalmology 2023A leading cause of irreversible vision loss, glaucoma needs early detection for effective management. Intraocular Pressure (IOP) is a significant risk factor for... (Review)
Review
A leading cause of irreversible vision loss, glaucoma needs early detection for effective management. Intraocular Pressure (IOP) is a significant risk factor for glaucoma. Convolutional Neural Networks (CNN) demonstrate exceptional capabilities in analyzing retinal fundus images, a non-invasive and cost-effective imaging technique widely used in glaucoma diagnosis. By learning from large datasets of annotated images, CNN can identify subtle changes in the optic nerve head and retinal structures indicative of glaucoma. This enables early and precise glaucoma diagnosis, empowering clinicians to implement timely interventions. CNNs excel in analyzing complex medical images, detecting subtle changes indicative of glaucoma with high precision. Another valuable diagnostic tool for glaucoma evaluation, Optical Coherence Tomography (OCT), provides high-resolution cross-sectional images of the retina. CNN can effectively analyze OCT scans and extract meaningful features, facilitating the identification of structural abnormalities associated with glaucoma. Visual field testing, performed using devices like the Humphrey Field Analyzer, is crucial for assessing functional vision loss in glaucoma. The integration of CNN with retinal fundus images, OCT scans, visual field testing, and IOP measurements represents a transformative approach to glaucoma detection. These advanced technologies have the potential to revolutionize ophthalmology by enabling early detection, personalized management, and improved patient outcomes. CNNs facilitate remote expert opinions and enhance treatment monitoring. Overcoming challenges such as data scarcity and interpretability can optimize CNN utilization in glaucoma diagnosis. Measuring retinal nerve fiber layer thickness as a diagnostic marker proves valuable. CNN implementation reduces healthcare costs and improves access to quality eye care. Future research should focus on optimizing architectures and incorporating novel biomarkers. CNN integration in glaucoma detection revolutionizes ophthalmology, improving patient outcomes and access to care. This review paves the way for innovative CNN-based glaucoma detection methods. CNN = Convolutional Neural Networks, AI = Artificial Intelligence, IOP = Intraocular Pressure, OCT = Optical Coherence Tomography, CLSO = Confocal Scanning Laser Ophthalmoscopy, AUC-ROC = Area Under the Receiver Operating Characteristic Curve, RNFL = Retinal Nerve Fiber Layer, RNN = Recurrent Neural Networks, VF = Visual Field, AP = Average Precision, MD = Mean Defect, sLV = square-root of Loss Variance, NN = Neural Network, WHO = World Health Organization.
Topics: Humans; Artificial Intelligence; Ophthalmology; Glaucoma; Optic Disk; Neural Networks, Computer; Intraocular Pressure; Tomography, Optical Coherence; Vision Disorders
PubMed: 37876506
DOI: 10.22336/rjo.2023.39 -
BMJ Open Ophthalmology Jun 2022To describe a new method to estimate the frequency distribution of optic nerve disc area, using digital retinographic images.
OBJECTIVE
To describe a new method to estimate the frequency distribution of optic nerve disc area, using digital retinographic images.
METHODS AND ANALYSIS
We analysed 492 023 fundus images obtained with seven fundus cameras, mainly in Caucasian subjects. They were grouped by resolution and zoom. They were automatically segmented by identifying the inner edge of the Elschnig scleral ring. For this purpose, a neural network trained by deep learning previously described was used. The number of pixels contained within the segmentation and their frequency distribution were calculated. The results of each camera, using different number of images, were compared with the global results using the Kolmogorov-Smirnov test to confront frequency distributions.
RESULTS
The frequency distribution was non-Gaussian, more limited in small sizes than in large ones. If the median is assigned a theoretical value of 1.95 mm, the 1th, 5th, 25th, 50th, 75th, 95th and 99th percentiles would correspond to 1.29, 1.46, 1.73, 1.95, 2.20, 2.64 and 3.03 mm in all the dataset. The overall differences were significant for the smaller series, but for each percentile their mean value was only 0.01 mm and the maximum 0.10 mm, so they can be considered similar for practical purposes in all cameras.
CONCLUSION
By automatically segmenting the edges of the optic nerve and observing the frequency distribution of the number of pixels it delimits, it is possible to estimate the frequency distribution of the disc area in the population as a whole and that of each individual case.
Topics: Diagnostic Techniques, Ophthalmological; Fundus Oculi; Glaucoma; Humans; Optic Disk; Optic Nerve
PubMed: 36161846
DOI: 10.1136/bmjophth-2022-000972 -
Molecular & Cellular Proteomics : MCP Nov 2023The pathogenesis of glaucoma is still unknown. There are few studies on the dynamic change of tissue-specific and time-specific molecular pathophysiology caused by...
The pathogenesis of glaucoma is still unknown. There are few studies on the dynamic change of tissue-specific and time-specific molecular pathophysiology caused by ocular hypertension (OHT). This study aimed to identify the early proteomic alterations in the retina, optic nerve head (ONH), and optic nerve (ON). After establishing a rat model of OHT, we harvested the tissues from control and glaucomatous eyes and analyzed the changes in protein expression using a multiplexed quantitative proteomics approach (TMT-MS3). Our study identified 6403 proteins after 1-day OHT and 4399 proteins after 7-days OHT in the retina, 5493 proteins after 1-day OHT and 4544 proteins after 7-days OHT in ONH, and 5455 proteins after 1-day OHT and 3835 proteins after 7-days OHT in the ON. Of these, 560 and 489 differential proteins were identified on day 1 and 7 after OHT in the retina, 428 and 761 differential proteins were identified on day 1 and 7 after OHT in the ONH, and 257 and 205 differential proteins on days 1 and 7 after OHT in the ON. Computational analysis on day 1 and 7 of OHT revealed that alpha-2 macroglobulin was upregulated across two time points and three tissues stably. The differentially expressed proteins between day 1 and 7 after OHT in the retina, ONH, and ON were associated with glutathione metabolism, mitochondrial dysfunction/oxidative phosphorylation, oxidative stress, microtubule, and crystallin. And the most significant change in retina are crystallins. We validated this proteomic result with the Western blot of crystallin proteins and found that upregulated on day 1 but recovered on day 7 after OHT, which are promising as therapeutic targets. These findings provide insights into the time- and region-order mechanisms that are specifically affected in the retina, ONH, and ON in response to elevated IOP during the early stages.
Topics: Rats; Animals; Optic Disk; Proteomics; Intraocular Pressure; Glaucoma; Retina; Ocular Hypertension; Optic Nerve; Crystallins
PubMed: 37793503
DOI: 10.1016/j.mcpro.2023.100654 -
Current Eye Research May 2023The optic nerve mechanically loads the eye during ocular rotation, thus altering the configuration of the disk and peripapillary tissues. We used optical coherence...
PURPOSE
The optic nerve mechanically loads the eye during ocular rotation, thus altering the configuration of the disk and peripapillary tissues. We used optical coherence tomography (OCT) angiography (OCTA) to investigate mechanical strains and volume changes in disk and peripapillary blood vessels during horizontal duction.
METHODS
Structural OCT and OCTA were performed centered on optic disks; imaging was repeated in central gaze, and in 30° ab- and adduction. By an algorithm employing point-set registration of 3 D features, we developed a novel approach for measuring disk strains, and strains and volumes of the blood vessels associated with horizontal duction. Repeatability was demonstrated in each gaze position.
RESULTS
19 eyes of 10 healthy adults of average age 37 ± 15 (standard deviation, SD) years were imaged. The method was validated by demonstrating numerically consistent vascular volumes and strains for repeated imaging under identical conditions. Compared with central gaze, vascular volume increased by 5.2 ± 4.1% in adduction. Adduction and abduction caused strains of 3.0 ± 1.6% and 2.6 ± 1.8% in the optic disk, whereas blood vessels showed greater strains of 8.1 ± 1.3% and 8.2 ± 1.7%. Decomposition of strain components depending on directionality and regions demonstrated that adduction induces significant net tensile strains, suggesting traction exerted by the optic nerve. The decomposition also showed that nasotemporal compressive strains are larger in temporal hemidisks than nasal hemidisks. The Bruch's membrane opening was significantly compressed horizontally in adduction by 1.1% ( = .009).
CONCLUSION
This novel analysis combining structural OCT and OCTA demonstrates that optic disk compression during adduction is associated with disk and vascular strains much larger than reported for intraocular pressure elevation and pulsatile perfusion, as well as compressing the disk and increasing peripapillary vascular volume. These changes may be relevant to the pathogenesis of optic nerve and retinal vascular disorders.
Topics: Adult; Humans; Young Adult; Middle Aged; Optic Disk; Tomography, Optical Coherence; Optic Nerve Diseases; Optic Nerve; Angiography; Intraocular Pressure; Retinal Vessels; Fluorescein Angiography
PubMed: 36843550
DOI: 10.1080/02713683.2023.2172185 -
BMC Ophthalmology May 2023To investigate morphological and microcirculation changes of optic nerve head (ONH) in simple high myopia (SHM) and pathologic myopia(PM) to evaluate and identify ONH...
PURPOSE
To investigate morphological and microcirculation changes of optic nerve head (ONH) in simple high myopia (SHM) and pathologic myopia(PM) to evaluate and identify ONH changes in the development of PM.
METHODS
A cross-sectional clinical study was used. Medical records from 193 right eyes of 193 patients with high myopia (HM) were included. Using the Topocon swept source optical coherence tomograph (SS-OCT) and fundus camera to detect the parameters, we have assessed the relative position and size of ONH, tilt and rotation of ONH, angle α (Defined as between retinal temporal arterial vascular arcades was measured from the centre of ONH with 250 pixels' radius), size and type of peripapillary atrophy (PPA), the thickness of peripapillary retinal nerve fiber layer (PRNFL), peripapillary choriodal thickness (PCT) and peripapillary scleral thickness (PST), and peripapillary vessel density (PVD). In addition, subjects were grouped as SHM and PM according to retinopathy, and the above parameters were compared between the two groups.
RESULTS
Patients were divided into the SHM group (138 eyes) and the PM group (55 eyes). Paramters like older age, higher diopter and longer axial length (AL) of the PM were compared to SHM (t=-3.585, -8.808, -11.409, all P<0.05). There were no differences in the smallest diameter and area of ONH, rotation angle and ratio, or PST (all P>0.05). The angle α in PM was smaller than that in SHM (t = 2.728, P<0.01). The disc-fovea distance (DFD), the largest diameter, tilt index and ratio, PPA area and radian in PM were larger than in SHM (t=-3.962, Z=-2.525, t=-2.229, Z=-4.303, Z=-2.834, all P<0.05). The superior and inferior PRNFLs in PM were smaller than in SHM (t = 4.172, 4.263, all P<0.01). The temporoinferior PRNFL was the opposite (t=-2.421, P<0.01). The average PCT in PM (93.82 ± 29.96 μm) was smaller than in SHM (108.75 ± 30.70 μm) (P<0.05). The PVD in each direction of PM was smaller than that in SHM (t = 6.398, 4.196, 4.971, 3.267, 5.029, 5.653, 4.202, 5.146, 2.090, all P<0.05).
CONCLUSION
Compared with SHM, the PM patients were older, with higher diopter. Their AL and DFD were longer, the angle α was smaller, the tilt index was more extensive, the PPA area and radian were larger, PCT was generally thinner, and PVD was lower. When the PPA area was bigger than the ONH area, this already indicated the presence of PM. Based on these results, we suggest ophthalmologists and myopia patients pay more attention to ONH's morphology and microcirculation changes as there is a possibility that microcirculatory changes precede morphologic changes.
Topics: Humans; Optic Disk; Microcirculation; Cross-Sectional Studies; Myopia; Tomography, Optical Coherence
PubMed: 37165324
DOI: 10.1186/s12886-023-02949-7 -
International Journal of Molecular... May 2022Huntington's disease (HD) is a fatal neurodegenerative disorder caused by a polyglutamine expansion in the huntingtin protein. HD-related pathological remodelling has...
Huntington's disease (HD) is a fatal neurodegenerative disorder caused by a polyglutamine expansion in the huntingtin protein. HD-related pathological remodelling has been reported in HD mouse models and HD carriers. In this study, we studied structural abnormalities in the optic nerve by employing Spectral Domain Optical Coherence Tomography (SD-OCT) in pre-symptomatic HD carriers of Caucasian origin. Transmission Electron Microscopy (TEM) was used to investigate ultrastructural changes in the optic nerve of the well-established R6/2 mouse model at the symptomatic stage of the disease. We found that pre-symptomatic HD carriers displayed a significant reduction in the retinal nerve fibre layer (RNFL) thickness, including specific quadrants: superior, inferior and temporal, but not nasal. There were no other significant irregularities in the GCC layer, at the macula level and in the optic disc morphology. The ultrastructural analysis of the optic nerve in R6/2 mice revealed a significant thinning of the myelin sheaths, with a lamellar separation of the myelin, and a presence of myelonoid bodies. We also found a significant reduction in the thickness of myelin sheaths in peripheral nerves within the choroids area. Those ultrastructural abnormalities were also observed in HD photoreceptor cells that contained severely damaged membrane disks, with evident vacuolisation and swelling. Moreover, the outer segment of retinal layers showed a progressive disintegration. Our study explored structural changes of the optic nerve in pre- and clinical settings and opens new avenues for the potential development of biomarkers that would be of great interest in HD gene therapies.
Topics: Animals; Huntington Disease; Mice; Nerve Fibers; Optic Disk; Optic Nerve; Retina
PubMed: 35628260
DOI: 10.3390/ijms23105450 -
Journal of Glaucoma Jun 2020To compare disease severity between preperimetric primary open-angle glaucoma (POAG) patients with and without deep-layer microvasculature dropout. (Comparative Study)
Comparative Study
PURPOSE
To compare disease severity between preperimetric primary open-angle glaucoma (POAG) patients with and without deep-layer microvasculature dropout.
MATERIALS AND METHODS
Ninety-four eyes of 94 preperimetric POAG patients with β-zone parapapillary atrophy (βPPA) were categorized according to the presence of deep-layer microvasculature dropout defined as a complete loss of microvasculature within the choroid or scleral flange on optical coherence tomography angiography. Parameters representing disease severity, that is, visual field (VF) mean deviation (MD), global and sectoral (6-sector) retinal nerve fiber layer (RNFL) thickness, and other factors including age, focal lamina cribrosa (LC) defect, width of βPPA with and without Bruch membrane (BM) (βPPA+BM and βPPA-BM), and optic disc hemorrhage were compared between eyes with and without dropout.
RESULTS
Deep-layer microvasculature dropout was observed in 33 preperimetric POAG eyes (35.1%). Eyes with dropout had significantly thinner RNFL in all areas except the inferonasal sector, worse VF MD, and higher prevalence of focal LC defect, and larger βPPA-BM (P<0.05), whereas the 2 groups did not differ in age, disc hemorrhage, or βPPA+BM width (P>0.05). In the multivariable logistic regression, worse VF MD [odds ratio (OR), 1.485; P=0.045], thinner RNFL (OR, 1.141; P<0.001), and higher prevalence of focal LC defect (OR, 6.673; P<0.001) were significantly associated with dropout.
CONCLUSIONS
Deep-layer microvasculature dropout was observed in a considerable number of preperimetric POAG eyes, and worse disease severity was associated with dropout. Future studies elucidating the pathogenic role of deep-layer microvasculature dropout in the development and progression of glaucoma are warranted.
Topics: Adult; Aged; Choroid; Cohort Studies; Disease Progression; Female; Glaucoma, Open-Angle; Humans; Intraocular Pressure; Male; Microvessels; Middle Aged; Optic Disk; Optic Nerve Diseases; Sclera; Severity of Illness Index; Tomography, Optical Coherence; Visual Fields
PubMed: 32205833
DOI: 10.1097/IJG.0000000000001489 -
Arquivos Brasileiros de Oftalmologia 2023The purpose of this study was to assess the optic nerve head microvascular changes in pseudoexfoliative and primary open-angle glaucoma and define the relationship... (Observational Study)
Observational Study
PURPOSE
The purpose of this study was to assess the optic nerve head microvascular changes in pseudoexfoliative and primary open-angle glaucoma and define the relationship between vessel density and retinal nerve fiber layer thickness.
METHODS
This observational cross-sectional study assessed 72 eyes with primary open-angle glaucoma, 41 eyes with pseudoexfoliative glaucoma, and 60 healthy eyes. On the basis of optic nerve head-centered, 4.5 mm × 4.5 mm scan size images, we evaluated the vessel density, as well as the peripapillary sector, inside disk, and all sectoral quadrants.
RESULTS
Both glaucoma Groups had lower vessel density in all regions compared with the healthy Group (p<0.05 for all variables). Vessel densities of the nasal inferior, inferior nasal, and inferior temporal sectors in both glaucoma Groups showed similar results (p=0.157, p=0.128, p=0.143, respectively). Eyes with pseudoexfoliative glaucoma had significantly lower vessel densities than eyes with primary open-angle glaucoma in all other regions (p<0.05 for all variables). For both glaucoma Groups, the average retinal nerve fiber layer thickness positively correlated with vessel density in all peripapillary sectors (p<0.05 for all variables).
CONCLUSIONS
Reduction in vessel density correlated with the thinning of retinal nerve fiber layer in both glaucoma Groups. Decreased vessel density in the optic nerve head can be used to demonstrate the microvascular pathologies and possible ischemic changes that lead to faster progression and worse prognosis in pseudoexfoliative glaucoma.
Topics: Humans; Optic Disk; Glaucoma, Open-Angle; Visual Fields; Tomography, Optical Coherence; Retinal Ganglion Cells; Nerve Fibers; Glaucoma; Retinal Vessels; Intraocular Pressure
PubMed: 37878951
DOI: 10.5935/0004-2749.2021-0420 -
Graefe's Archive For Clinical and... May 2022To assess the optic nerve pial diameter (ONPD) in patients with primary open-angle glaucoma (POAG) using standardized A-scan ultrasound and to evaluate the correlation...
PURPOSE
To assess the optic nerve pial diameter (ONPD) in patients with primary open-angle glaucoma (POAG) using standardized A-scan ultrasound and to evaluate the correlation between the ONPD and structural, vascular optic nerve head features and visual field parameters in glaucomatous eyes.
METHODS
In this prospective study, we enrolled 126 eyes of 63 POAG patients and 124 eyes of 62 healthy controls. In all subjects, the ONPD was evaluated by means of A-scan ultrasound. Spectral domain (SD)-OCT was used to assess ganglion cell complex (GCC), retinal nerve fiber layer (RNFL), thicknesses, and the optic nerve head (ONH) morphology. OCTA measured the vessel density (VD) of radial peripapillary capillary (RPC) plexus.
RESULTS
The ONPD showed a statistically significant reduction in POAG group with respect to controls (p < 0.001). SD-OCT and OCTA parameters showed a significant impairment in patient group with respect to controls (p < 0.001). The ONH analysis revealed significantly lower values in rim area (p = 0.009) and an increased cup-to-disc area ratio (p = 0.013) and cup volume (p < 0.001) in patients with respect to controls. Significant correlations were shown in POAG group between ONPD and RPC plexus (p = 0.006). Moreover, significant correlation was also found between ONPD and structural SD-OCT parameters (p = 0.001) and between ONPD and visual field parameters (p = 0.001).
CONCLUSIONS
The standardized A-scan ultrasound measurements of the ONPD showed a significant correlation with structural and vascular glaucomatous changes measured by means of SD-OCT and OCTA. These results confirm the diagnostic reliability of the ultrasound evaluation in glaucoma optic neuropathy.
Topics: Glaucoma, Open-Angle; Humans; Intraocular Pressure; Optic Disk; Prospective Studies; Reproducibility of Results; Retinal Vessels; Tomography, Optical Coherence
PubMed: 34674028
DOI: 10.1007/s00417-021-05438-z