Estimating the Amount of Hemoglobin in the Neuroretinal Rim Using Color Images and OCT

Medina-Mesa, Erica; González-Hernández, Marta; Sigut, J.; Fumero-Batista, Francisco; Pena-Betancor, Cristina; Alayón, Silvia; Rosa, Manuel González de la

doi:10.3109/02713683.2015.1062112

Cited by 17 publications

(28 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A total of 2357 images from five public glaucoma-labelled databases: HRF [8] (45 images), Drishti-GS1 [9] (101 images), ORIGA-light [10] (650 images) RIM-ONE [11] (455 images) and sjchoi86-HRF [12] database (401 images) and a private database, ACRIMA (705 images), were used to train the generative models used in this work. All images were manually cropped around the optic disc by an expert, with the exception of RIM-ONE images that are already cropped.…”

Section: Methodsmentioning

confidence: 99%

Retinal Image Synthesis for Glaucoma Assessment Using DCGAN and VAE Models

Diaz-Pinto

Colomer

Naranjo

et al. 2018

Intelligent Data Engineering and Automated Learning – IDEAL 2018

View full text Add to dashboard Cite

The performance of a Glaucoma assessment system is highly affected by the number of labelled images used during the training stage. However, labelled images are often scarce or costly to obtain. In this paper, we address the problem of synthesising retinal fundus images by training a Variational Autoencoder and an adversarial model on 2357 retinal images. The innovation of this approach is in synthesising retinal images without using previous vessel segmentation from a separate method, which makes this system completely independent. The obtained models are image synthesizers capable of generating any amount of cropped retinal images from a simple normal distribution. Furthermore, more images were used for training than any other work in the literature. Synthetic images were qualitatively evaluated by 10 clinical experts and their consistency were estimated by measuring the proportion of pixels corresponding to the anatomical structures around the optic disc. Moreover, we calculated the mean-squared error between the average 2D-histogram of synthetic and real images, obtaining a small difference of 3 × 10 −4. Further analysis of the latent space and cup size of the images was performed by measuring the Cup/Disc ratio of synthetic images using a state-of-the-art method. The results obtained from this analysis and the qualitative and quantitative evaluation demonstrate that the synthesised images are anatomically consistent and the system is a promising step towards a model capable of generating labelled images.

show abstract

Section: Methodsmentioning

confidence: 99%

Retinal Image Synthesis for Glaucoma Assessment Using DCGAN and VAE Models

Diaz-Pinto

Colomer

Naranjo

et al. 2018

Intelligent Data Engineering and Automated Learning – IDEAL 2018

View full text Add to dashboard Cite

show abstract

“…In order to analyze the results in this way, the colocation of the rim sectors with respect to OCT is very important [11, 12], but this was not our intention in the present work. In the progression of glaucoma, the cup grows at the expense of rim reduction in the intermediate zones of the ONH.…”

Section: Discussionmentioning

confidence: 99%

“…However, the automatic edge identification algorithms used by this program take into account the experience accumulated in other studies in which the photographic images were adjusted with the ONH edges delimited by OCT [12]. …”

Section: Methodsmentioning

confidence: 99%

“…New studies have allowed separating the rim from cup information, analyzing images obtained with a stereoscopic camera [11], and also associating two-dimensional photographic images with optical disk and cup delimitations provided by OCT [12]. Among the conclusions of these studies, two stand out: first, hemoglobin concentrations are better correlated with functional defects of glaucoma than the corresponding areas of the neuroretinal rim or RNFL thicknesses.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Relationship between Retinal Nerve Fiber Layer Thickness and Hemoglobin Present in the Optic Nerve Head in Glaucoma

González-Hernández

Sigut

Rosa

2017

Journal of Ophthalmology

Self Cite

View full text Add to dashboard Cite

Purpose To observe the relationship between topographic hemoglobin levels in the optic nerve head (ONH), the rim thickness (BMO-MRW), and retinal nerve fiber layer (RNFL) thickness. Methods 96 normal eyes and 82 glaucomas were examined using TOP strategy (Octopus 300 perimeter), SPECTRALIS OCT, and Laguna ONhE program which estimates hemoglobin from conventional color photographs (Horus Scope DEC 200 fundus camera). Results The correlation between Laguna ONhE glaucoma discriminant function (GDF) and SPECTRALIS BMO-MRW was R = 0.81 (P < 0.0001), similar to that between the BMO-MRW and BMO-RNFL thicknesses (R = 0.85, P < 0.0001) (P = 0.227 between both R values). GDF correlated well with RNFL thicknesses in the 360 degrees around the nerve, similar to mean perimetric sensitivity (MS) and BMO-MRW. The amount of hemoglobin in the nasal and temporal sectors showed low correlation with superior and inferior RNFL thicknesses. The superotemporal and inferotemporal sectors located on the vertical diameter of the disk showed good intercorrelation but without a clear RNFL topographic relationship. Conclusion GDF showed high correlation with RNFL thickness. Except in the nasal and temporal sectors, ONH hemoglobin correlated well with RNFL thickness.

show abstract

“…HRF database, which contains 45 images [16]. RIM-ONE database, which is composed of 455 images [17], and Autogla database, which is composed of 83 images provided by the FISABIO oftalmología médica (FOM). Experts of the FISABIO Oftalmología Médica (FOM) analysed the images of the DRIVE database for glaucoma through visual inspection, and based on their analysis they were labelled as Glaucoma or Normal.…”

Section: Methodsmentioning

confidence: 99%

Computer-Aided Glaucoma Diagnosis Using Stochastic Watershed Transformation on Single Fundus Images

Diaz-Pinto¹,

Morales²,

Naranjo³

et al. 2019

j med imaging hlth inform

View full text Add to dashboard Cite

Glaucoma is a chronic eye disease and one of the major causes of permanent blindness. Since it does not show initial symptoms, early diagnosis is important to limit its progression. This paper presents an automatic optic nerve characterization algorithm for glaucoma diagnosis based only on retinal fundus images. For optic cup segmentation, we used a new method based on the stochastic watershed transformation applied on the YIQ colour space to extract clinical indicators such as the Cup/Disc ratio, the area Cup/Disc ratio and the ISNT rule. Afterwards, an assessment between normal and glaucomatous fundus images is performed. The proposed algorithm was evaluated on 6 different (private and public) databases containing 723 images (377 normal and 346 glaucomatous images) which achieved a specificity and sensitivity of 0.674 and 0.675, respectively. Moreover, an F-score of 0.770 was obtained when evaluating this method on the publicly available database Drishti-GS1. A comparison of the proposed work with other state-of-the-art methods demonstrates the robustness of the proposed algorithm; because it was tested using images from different databases with high variability, which is a common issue in this area. Additional comparisons with existing works for cup segmentation, that use the publicly available database Drishti-GS1, are also presented in this paper.

show abstract

Estimating the Amount of Hemoglobin in the Neuroretinal Rim Using Color Images and OCT

Abstract: The amount of Hb in the ONH seems to have an important relationship with glaucomatous visual field sensitivity. The remaining rim has insufficient perfusion in many cases of glaucoma.

Cited by 17 publications

References 30 publications

Retinal Image Synthesis for Glaucoma Assessment Using DCGAN and VAE Models

Retinal Image Synthesis for Glaucoma Assessment Using DCGAN and VAE Models

Relationship between Retinal Nerve Fiber Layer Thickness and Hemoglobin Present in the Optic Nerve Head in Glaucoma

Computer-Aided Glaucoma Diagnosis Using Stochastic Watershed Transformation on Single Fundus Images

Contact Info

Product

Resources

About