Exudate-based diabetic macular edema detection in fundus images using publicly available datasets

Giancardo, Luca; Mériaudeau, Fabrice; Karnowski, Thomas P.; Li, Yaquin; Garg, Seema; Tobin, Kenneth W.; Chaum, Edward

doi:10.1016/j.media.2011.07.004

Cited by 287 publications

(175 citation statements)

References 33 publications

Supporting

Mentioning

164

Contrasting

Unclassified

Order By: Relevance

“…In traditional automated grading systems [5] - [10], in order to grade DME, the contrast of images is enhanced as a pre-processing stage and then the blood vessels are removed using matched filtering or mathematical morphology. Further to this, the location of the macula is detected and exudate segmentation is applied.…”

Section: System Overviewmentioning

confidence: 99%

“…Furthermore, Zaidi et al [7] developed a grading method using Gabor filtering, mathematical morphology and Otsu thresholding with a Bayesian classifier to detect the location of exudate and positional constraints to grade the severity of DME. Moreover, Giancardo et al [5] proposed an automated grading system based on exudate probability map and wavelet decomposition. The Kirsch edge operator and a region-growing algorithm were used to locate the hard exudate and the fovea region.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Diabetic Macular Edema Grading Based on Deep Neural Networks

Al-Bander

Al-Nuaimy

Al-Taee

et al. 2016

Proceedings of the Ophthalmic Medical Image Analysis Third International Workshop

View full text Add to dashboard Cite

Abstract. Diabetic Macular Edema (DME) is a major cause of vision loss in diabetes. Its early detection and treatment is therefore a vital task in management of diabetic retinopathy. In this paper, we propose a new featurelearning approach for grading the severity of DME using color retinal fundus images. An automated DME diagnosis system based on the proposed featurelearning approach is developed to help early diagnosis of the disease and thus averts (or delays) its progression. It utilizes the convolutional neural networks (CNNs) to identify and extract features of DME automatically without any kind of user intervention. The developed prototype was trained and assessed by using an existing MESSIDOR dataset of 1200 images. The obtained preliminary results showed accuracy of (88.8 %), sensitivity (74.7%) and specificity (96.5 %). These results compare favorably to state-of-the-art findings with the added benefit of an automatic feature-learning approach rather than a time-consuming handcrafted approach.

show abstract

Section: System Overviewmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Diabetic Macular Edema Grading Based on Deep Neural Networks

Al-Bander

Al-Nuaimy

Al-Taee

et al. 2016

Proceedings of the Ophthalmic Medical Image Analysis Third International Workshop

View full text Add to dashboard Cite

show abstract

“…Some studies have been done with different algorithms for automatic DME assessment. Giancardo et al [11], used a feature vector representing the exudate probability map along with a wavelet decomposition and classifier for for automatic lesion segmentation and DME assessment. Vasanthi and Banu [12]studied the preprocessing of images and the use of an adaptive neuro-fuzzy inference system and an extreme learning machine to classify eye fundus images between normal and with some kind of DME.…”

Section: Introductionmentioning

confidence: 99%

A Novel Machine Learning Model Based on Exudate Localization to Detect Diabetic Macular Edema

Perdómo

Otálora

Rodríguez³

et al. 2016

Proceedings of the Ophthalmic Medical Image Analysis Third International Workshop

View full text Add to dashboard Cite

Diabetic macular edema is one of the leading causes of legal blindness worldwide. Early, and accessible, detection of ophthalmological diseases is especially important in developing countries, where there are major limitations to access to specialized medical diagnosis and treatment. Deep learning models, such as deep convolutional neural networks have shown great success in different computer vision tasks. In medical images they have been also applied with great success. The present paper presents a novel strategy based on convolutional neural networks to combine exudates localization and eye fundus images for automatic classification of diabetic macular edema as a support for diabetic retinopathy diagnosis.

show abstract

“…Without adequate treatment, it can evolve into a more complicated condition called Macular Edema [11]. Typically, the procedures used for retinal analysis include the identification of ocular physiology elements, like blood vessels, optic disc, and fovea [12][13].…”

Section: Introductionmentioning

confidence: 99%

A GPU-based Evolution Strategy for Optic Disk Detection in Retinal Images

Sánchez-Torres

González-Calederón

2016

rev.ing.univ.Medellin

View full text Add to dashboard Cite

Parallel processing using graphic processing units (gpus) has attracted much research interest in recent years. Parallel computation can be applied to evolution strategy (es) for processing individuals in a population, but evolutionary strategies are time-consuming when used to solve large computational problems or complex fitness functions. In this paper, we describe the implementation of an improved es for optic disk detection in retinal images using the Compute Unified Device Architecture (cuda) environment. In the experimental results, we show that the computational time for optic disk detection task has a speedup factor of 5x and 7x compared to the implementation on a mainstream cpu. Key words:Compute Unified Device Architecture, optic disk, evolutionary strategy, retinal images. Revista Ingenierías Universidad de MedellínGermán Sanchez-Torres -Guillermo González-Calderón Estrategia evolutiva basada en GPU para la detección del disco óptico en imágenes de retina Resumen La ejecución paralela de aplicaciones usando unidades de procesamiento gráfico (gpu) ha ganado gran interés en la comunidad académica en los años recientes. La computación paralela puede ser aplicada a las estrategias evolutivas para procesar individuos dentro de una población, sin embargo, las estrategias evolutivas se caracterizan por un significativo consumo de recursos computacionales al resolver problemas de gran tamaño o aquellos que se modelan mediante funciones de aptitud complejas. Este artículo describe la implementación de una estrategia evolutiva para la detección del disco óptico en imágenes de retina usando Compute Unified Device Architecture (cuda). Los resultados experimentales muestran que el tiempo de ejecución para la detección del disco óptico logra una aceleración de 5 a 7 veces, comparado con la ejecución secuencial en una cpu convencional.Palabras clave: gpu, disco óptico, estrategias evolutivas, imágenes de retina.

show abstract

Exudate-based diabetic macular edema detection in fundus images using publicly available datasets

Cited by 287 publications

References 33 publications

Diabetic Macular Edema Grading Based on Deep Neural Networks

Diabetic Macular Edema Grading Based on Deep Neural Networks

A Novel Machine Learning Model Based on Exudate Localization to Detect Diabetic Macular Edema

A GPU-based Evolution Strategy for Optic Disk Detection in Retinal Images

Contact Info

Product

Resources

About