Sirius: A web-based system for retinal image analysis

Barreira, N.; Novo, Jorge; Penedo, Manuel G.; Pose‐Reino, Antonio; Gómez‐Ulla, Francisco

doi:10.1016/j.ijmedinf.2010.07.005

Cited by 50 publications

(28 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The line-strength of the target pixel was obtained from the orientation with the largest value subtracted by the average grey-level of a square sub-window centred at the target pixel. This was a multi-scale approach, so lines of different lengths (15,17,19,21 pixels) were applied and a linear combination of responses resulted in a final line-strength image. The application of a hysteresis thresholding based morphological reconstruction then followed to produce the segmented vasculature.…”

Section: Vessel Segmentationmentioning

confidence: 99%

Automated retinal image quality assessment on the UK Biobank dataset for epidemiological studies

Welikala

Fraz

Foster

et al. 2016

Computers in Biology and Medicine

View full text Add to dashboard Cite

Abstract:Morphological changes in the retinal vascular network are associated with future risk of many systemic and vascular diseases. However, uncertainty over the presence and nature of some of these associations exists. Analysis of data from large population based studies will help to resolve these uncertainties. The QUARTZ (QUantitative Analysis of Retinal vessel Topology and siZe) retinal image analysis system allows automated processing of large numbers of retinal images. However, an image quality assessment module is needed to achieve full automation. In this paper, we propose such an algorithm, which uses the segmented vessel map to determine the suitability of retinal images for use in the creation of vessel morphometric data suitable for epidemiological studies. This includes an effective 3-dimensional feature set and support vector machine classification. A random subset of 800 retinal images from UK Biobank (a large prospective study of 500,000 middle aged adults; where 68,151 underwent retinal imaging) was used to examine the performance of the image quality algorithm. The algorithm achieved a sensitivity of 95.33% and a specificity of 91.13%for the detection of inadequate images. The strong performance of this image quality algorithm will make rapid automated analysis of vascular morphometry feasible on the entire UK Biobank dataset (and other large retinal datasets), with minimal operator involvement, and at low cost.

show abstract

Section: Vessel Segmentationmentioning

confidence: 99%

Automated retinal image quality assessment on the UK Biobank dataset for epidemiological studies

Welikala

Fraz

Foster

et al. 2016

Computers in Biology and Medicine

View full text Add to dashboard Cite

show abstract

“…"Retinal Analysis Tool" (ADCIS) includes semi-automatic tools for vessel segmentation but focuses on diabetic retinopathy and age-related macular degeneration. "SIRIUS" [12] is an online-system for lesion detection and AVR computation. "DR-CAD" [1] is a screening system for diabetes that uses machine learning to detect vessels, optic disk, and lesions.…”

Section: Related Workmentioning

confidence: 99%

An Image Analysis System for the Assessment of Retinal Microcirculation in Hypertension and Its Clinical Evaluation

et al. 2014

View full text Add to dashboard Cite

Abstract-A system for the assessment of hypertension through the measurement of retinal vessels in fundoscopy images, is presented. The proposed approach employs multiple image analysis methods, in an integrated system that is used in clinical practice. Automating the measurement process enables the conduct of a clinical study that, for the first time, shows the correlation between macrovascular and microvascular alterations, based on numerous measurements acquired by this system. Experience and perspectives gained from clinical usage and evaluation are reported.

show abstract

“…In the semiautomatic approach [6], once the medical experts had classified the vessels into arteries and veins, the AVR is computed as the ratio between the artery and vein average widths. However, the experts do not take into account all the detected vessels but only a subset of them.…”

Section: Vessel Selectionmentioning

confidence: 99%

“…We had developed a web-based system [6] for retinal image analysis which includes a service to compute the AVR in a semiautomatic way. In that system, once the experts have selected the optic disc position, the system detects the vessels in several circumferences centered at the optic nerve and measures their widths.…”

Section: Introductionmentioning

confidence: 99%

Automatic Arteriovenous Ratio Computation: Emulating the Experts

Vázquez

Barreira

Penedo

et al. 2012

Technological Innovation for Value Creation

Self Cite

View full text Add to dashboard Cite

Abstract. The arteriovenous ratio is an objective way to assess the arteriolar narrowing related to several diseases such as hypertension. It is computed as the ratio between the artery and vein mean widths. However, its calculus is not straightforward since the experts do not use all the retinal vessels. This paper presents an automatic, precise and reproducible methodology for the AVR computation. We analyze the way the experts select the vessels in order to build a system which emulates them. The system was evaluated by two ophthalmologists in a data set of 86 images. The correlation results among the system and the experts are an indication of the reproducibility of the results.

show abstract

Sirius: A web-based system for retinal image analysis

Cited by 50 publications

References 18 publications

Automated retinal image quality assessment on the UK Biobank dataset for epidemiological studies

Automated retinal image quality assessment on the UK Biobank dataset for epidemiological studies

An Image Analysis System for the Assessment of Retinal Microcirculation in Hypertension and Its Clinical Evaluation

Automatic Arteriovenous Ratio Computation: Emulating the Experts

Contact Info

Product

Resources

About