Towards a Glaucoma Risk Index Based on Simulated Hemodynamics from Fundus Images

Orlando, José Ignacio; Breda, João Barbosa; Keer, Karel Van; Blaschko, Matthew B.; Blanco, Pablo J.; Bulant, Carlos A.

doi:10.1007/978-3-030-00934-2_8

Cited by 61 publications

(35 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Regarding clinical relevance, a recent study on diabetic retinopathy showed that a combination of fundus and OCTA measurements can already improve assessment of the disease (130). Another study used a fundus-based model to generate a fixed length feature vector as an index that can be used to successfully discriminate between healthy and glaucoma (96). However, since they used constant, nonindividualized inlet values and relied on the vessel structure to guide the outcome, the end point parameters are not easily interpretable.…”

Section: Physiological Implications and Clinical Relevancementioning

confidence: 99%

Microcirculatory model predicts blood flow and autoregulation range in the human retina: in vivo investigation with laser speckle flowgraphy

Pappelis

Choritz

Jansonius

2020

American Journal of Physiology-Heart and Circulatory Physiology

View full text Add to dashboard Cite

In this study, 1) we mathematically predict Retinal Vascular Resistance (RVR) and Retinal Blood Flow (RBF), 2) we test predictions using Laser Speckle Flowgraphy (LSFG), 3) we estimate the range of vascular autoregulation, and 4) we examine the relationship of RBF with the retinal nerve fiber layer (RNFL) and ganglion cell complex (GCC). Fundus, OCT, and OCT-angiography images, systolic/diastolic blood pressure (SBP/DBP), and intraocular pressure (IOP) measurements were obtained from 36 human subjects. We modeled two circulation markers (RVR and RBF) and estimated individualized lower/higher autoregulation limits (LARL/HARL), using retinal vessel calibers, fractal dimension, perfusion pressure, and population-based hematocrit values. Quantitative LSFG waveforms were extracted from vessels of the same eyes, before and during IOP elevation. LSFG metrics explained most variance in RVR (R2=0.77/P=6.9·10-9) and RBF (R2=0.65/P=1.0·10-6), suggesting that the markers strongly reflect blood flow physiology. Higher RBF was associated with thicker RNFL (P=4.0·10-4) and GCC (P=0.003), thus also verifying agreement with structural measurements. LARL was at SBP/DBP of 105/65 mmHg for the average subject without arterial hypertension, and at 115/75 mmHg for the average hypertensive subject. Moreover, during IOP elevation, changes in RBF were more pronounced than changes in RVR. These observations physiologically imply that healthy subjects are already close to LARL, thus prone to hypoperfusion. In conclusion, we modeled two clinical markers and described a novel method to predict individualized autoregulation limits. These findings could improve understanding of retinal perfusion and pave the way for personalized intervention decisions, when treating patients with coexisting ophthalmic and cardiovascular pathologies.

show abstract

Section: Physiological Implications and Clinical Relevancementioning

confidence: 99%

Microcirculatory model predicts blood flow and autoregulation range in the human retina: in vivo investigation with laser speckle flowgraphy

Pappelis

Choritz

Jansonius

2020

American Journal of Physiology-Heart and Circulatory Physiology

View full text Add to dashboard Cite

show abstract

“…It also includes the disc and cup segmentation. DR HAGIS (Holm et al, 2017), HRF (Odstrcilik et al, 2013), and LES-AV (Orlando et al, 2018) are small sets with 39, 45 and 22 images, respectively, with no disc and cup segmentation. The ACRIMA set (Diaz-Pinto et al, 2019) contains 396 images from patients with glaucoma and 309 images from healthy patients.…”

Section: Related Workmentioning

confidence: 99%

RIM-ONE DL: A Unified Retinal Image Database for Assessing Glaucoma Using Deep Learning

et al. 2020

View full text Add to dashboard Cite

The first version of the Retinal IMage database for Optic Nerve Evaluation (RIM-ONE) was published in 2011. This was followed by two more, turning it into one of the most cited public retinography databases for evaluating glaucoma. Although it was initially intended to be a database with reference images for segmenting the optic disc, in recent years we have observed that its use has been more oriented toward training and testing deep learning models. The recent REFUGE challenge laid out some criteria that a set of images of these characteristics must satisfy to be used as a standard reference for validating deep learning methods that rely on the use of these data. This, combined with the certain confusion and even improper use observed in some cases of the three versions published, led us to consider revising and combining them into a new, publicly available version called RIM-ONE DL (RIM-ONE for Deep Learning). This paper describes this set of images, consisting of 313 retinographies from normal subjects and 172 retinographies from patients with glaucoma. All of these images have been assessed by two experts and include a manual segmentation of the disc and cup. It also describes an evaluation benchmark with different models of well-known convolutional neural networks.

show abstract

“…In this section, the lists of datasets used for segmenting and classifying retina fundus images are discussed. This datasets are listed in order of usage, they include: DRIVE [125], STARE [126], CHASE-DB1 [127], HRF [128], MESSIDOR [129], IOSTAR [130], ORIGA [131], REFUGE [131], DB1 [132], DB0 [133], Kaggle dataset, DRISHTI-GS [134], NIVE [135], RIM-ONEr3 [136], DRIONS-DB [137], RITE [138], WIDE [139], SYNTHE [140], LES-AV [141], RIGA [142], DUKE, DCA, EIARG1, and AV-INSPIRE. The list of datasets that have the highest usage is shown in Fig.…”

Section: Datasetsmentioning

confidence: 99%

Segmentation and Classification of Retinal Fundus Images using Convolutional Neural Networks: A systematic review of methods and latest trends

Ilesanmi

Idowu

et al. 2022

Preprint

View full text Add to dashboard Cite

The retina fundus images are critical diagnostic tools for the early detection of ophthalmic problems. Prompt diagnosis and treatment of eye-related problems are important to prevent blindness or vision loss. Recently, CNN algorithms have become effective in tasks relating to recognition, delineation, and classification. Therefore we propose a review to summarize different CNN algorithms for segmenting and classifying retinal fundus images. This review systematically searches different repositories for methods that use CNN for the segmentation and classification of retinal fundus images. A thorough screening of abstracts and titles to ascertain their relevance was done. A total of fifty-two studies were included in our review, with content such as database usage, disadvantages, and advantages. A comparison of two accuracies was also carried out with a graph depicting database usage. Important insights, limitations, observations, and future directions were elucidated. Finally, findings suggest that CNN algorithms produce good accuracies despite their limitations.

show abstract

Towards a Glaucoma Risk Index Based on Simulated Hemodynamics from Fundus Images

Cited by 61 publications

References 19 publications

Microcirculatory model predicts blood flow and autoregulation range in the human retina: in vivo investigation with laser speckle flowgraphy

Microcirculatory model predicts blood flow and autoregulation range in the human retina: in vivo investigation with laser speckle flowgraphy

RIM-ONE DL: A Unified Retinal Image Database for Assessing Glaucoma Using Deep Learning

Segmentation and Classification of Retinal Fundus Images using Convolutional Neural Networks: A systematic review of methods and latest trends

Contact Info

Product

Resources

About