Segmentation of the Optic Nerve Head Based on Deep Learning to Determine its Hemoglobin Content in Normal and Glaucomatous Subjects

Gonzalez-Hernandez, Daniel; Díaz-Alemán, Tinguaro; Perez-Barbudo, Daniel; Méndez-Hernández, Carmen; Rosa, Manuel González de la; González-Hernández, Marta

doi:10.4172/2155-9570.1000760

Cited by 4 publications

(10 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To achieve full automation of the method, five neural networks were used: one to segment the edges of the optic nerve already described [ 13 ], one for vessel segmentation using 4195 optic disc images, one to identify the eye as left or right using 4201 images, one to recognize image quality using 7048 images, and one to identify between normality (using 1518 images) and confirmed or suspected glaucoma (using 1596 images). The technical method is described in detail in the Appendix A “Computing development setup”.…”

Section: Methodsmentioning

confidence: 99%

“…A. Optic Disc segmentation: In a previous experiment [ 13 ], we used a deep learning U-Net architecture [ 16 ] for semantic segmentation. In this case, 40,000 images were used to segment the optic disc, identifying the inner edge of Elschnig’s scleral ring.…”

Section: Figure A1mentioning

confidence: 99%

“…Elschnig’s scleral ring is observed as a thin white ring, immediately adjacent to the margin of the optic disc. An example is shown in Figure 1 of reference [ 13 ]. B.…”

Section: Figure A1mentioning

confidence: 99%

“…The method has been shown to have good sensitivity and specificity in previous studies [ 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. However, deep learning methods for the segmentation of optic nerve edges were later incorporated to facilitate its use and improve its reproducibility [ 13 ].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Fully Automated Colorimetric Analysis of the Optic Nerve Aided by Deep Learning and Its Association with Perimetry and OCT for the Study of Glaucoma

González-Hernández

Gonzalez-Hernandez

Perez-Barbudo

et al. 2021

JCM

Self Cite

View full text Add to dashboard Cite

Background: Laguna-ONhE is an application for the colorimetric analysis of optic nerve images, which topographically assesses the cup and the presence of haemoglobin. Its latest version has been fully automated with five deep learning models. In this paper, perimetry in combination with Laguna-ONhE or Cirrus-OCT was evaluated. Methods: The morphology and perfusion estimated by Laguna ONhE were compiled into a “Globin Distribution Function” (GDF). Visual field irregularity was measured with the usual pattern standard deviation (PSD) and the threshold coefficient of variation (TCV), which analyses its harmony without taking into account age-corrected values. In total, 477 normal eyes, 235 confirmed, and 98 suspected glaucoma cases were examined with Cirrus-OCT and different fundus cameras and perimeters. Results: The best Receiver Operating Characteristic (ROC) analysis results for confirmed and suspected glaucoma were obtained with the combination of GDF and TCV (AUC: 0.995 and 0.935, respectively. Sensitivities: 94.5% and 45.9%, respectively, for 99% specificity). The best combination of OCT and perimetry was obtained with the vertical cup/disc ratio and PSD (AUC: 0.988 and 0.847, respectively. Sensitivities: 84.7% and 18.4%, respectively, for 99% specificity). Conclusion: Using Laguna ONhE, morphology, perfusion, and function can be mutually enhanced with the methods described for the purpose of glaucoma assessment, providing early sensitivity.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Figure A1mentioning

confidence: 99%

“…Elschnig’s scleral ring is observed as a thin white ring, immediately adjacent to the margin of the optic disc. An example is shown in Figure 1 of reference [ 13 ]. B.…”

Section: Figure A1mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Fully Automated Colorimetric Analysis of the Optic Nerve Aided by Deep Learning and Its Association with Perimetry and OCT for the Study of Glaucoma

González-Hernández

Gonzalez-Hernandez

Perez-Barbudo

et al. 2021

JCM

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our group has developed a procedure to topographically estimate relative haemoglobin levels in the optic nerve, called Laguna ONhE, [16][17][18][19][20][21] which in its most recent versions uses several convolutional neural networks. 22 One of them performs an automatic segmentation of the optic disc trying to detect the inner edge of the Elschnig scleral ring. The procedure has been applied to multiple fundus cameras and a large number of subjects.…”

Section: How This Study Might Affect Research Practice or Policy?mentioning

confidence: 99%

Optic disc area frequency distribution in a large sample of retinographic images

et al. 2022

Self Cite

View full text Add to dashboard Cite

ObjectiveTo describe a new method to estimate the frequency distribution of optic nerve disc area, using digital retinographic images.Methods and analysisWe 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.ResultsThe 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 mm2, 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 mm2 in all the dataset. The overall differences were significant for the smaller series, but for each percentile their mean value was only 0.01 mm2 and the maximum 0.10 mm2, so they can be considered similar for practical purposes in all cameras.ConclusionBy 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.

show abstract

Discrimination Between Healthy Eyes and Those With Mild Glaucoma Damage Using Hemoglobin Measurements of the Optic Nerve Head

Meneses

Ciarlini

Ayub

et al. 2022

Journal of Glaucoma

View full text Add to dashboard Cite

The Laguna ONhE, a software that measures the hemoglobin (Hb) concentration of the optic nerve head (ONH) from fundus photographs, demonstrated good accuracy in discriminating healthy eyes from eyes with mild glaucoma. Purpose:The aim was to evaluate Hb concentration of the optic nerve to distinguish between healthy eyes and eyes with mild glaucoma.Methods: Eyes from patients with mild primary open angle glaucoma (MD > −6 dB) (n = 58) and from healthy subjects (n = 64) were selected. Retinal nerve fiber layer thickness measurements of all eyes were acquired with optical coherence tomography. Optic disc photographs were also obtained, and the images were analyzed using the Laguna ONhE software, which measures the amount of Hb in 24 sectors of the ONH. The software also calculates the Glaucoma Discriminant Function (GDF), an index that expresses the chance of the ONH being compatible with glaucoma. Areas under the receiver operating characteristic curve and sensitivities at fixed specificities of 90% and 95% of each Laguna ONhE parameter were calculated. Results:The mean retinal nerve fiber layer thickness and vertical cup/disc ratio of the control and glaucoma groups were 90.0 ± 10.6 μm versus 66.28 ± 9.85 μm (P < 0.001) and 0.5 ± 0.09 versus 0.65 ± 0.09 (P < 0.001), respectively. Total Hb (67.9 ± 4.45 vs. 62.89 ± 4.89, P < 0.001) and , P < 0.001) were significantly higher in the control group. The Hb concentration was also significantly higher in 21 of the 24 sectors in the control group compared with the glaucoma group (P < 0.05). The GDF had the largest areas under the receiver operating characteristic curve (0.93), with 79.3% sensitivity at a fixed specificity of 95%. Conclusion:Measurements of optic nerve Hb concentration using a colorimetry photographic device demonstrated good accuracy in discriminating healthy eyes from eyes with mild glaucoma. Further studies are need to understand vascular factors implicated in the development of glaucoma.

show abstract

Segmentation of the Optic Nerve Head Based on Deep Learning to Determine its Hemoglobin Content in Normal and Glaucomatous Subjects

Cited by 4 publications

References 23 publications

Fully Automated Colorimetric Analysis of the Optic Nerve Aided by Deep Learning and Its Association with Perimetry and OCT for the Study of Glaucoma

Fully Automated Colorimetric Analysis of the Optic Nerve Aided by Deep Learning and Its Association with Perimetry and OCT for the Study of Glaucoma

Optic disc area frequency distribution in a large sample of retinographic images

Discrimination Between Healthy Eyes and Those With Mild Glaucoma Damage Using Hemoglobin Measurements of the Optic Nerve Head

Contact Info

Product

Resources

About