Automatic CDR Estimation for Early Glaucoma Diagnosis.

Glaucoma is a degenerative disease that constitutes the second cause of blindness in developed countries. Although it cannot be cured, its progression can be prevented through early diagnosis. In this paper, we propose a new algorithm for automatic glaucoma diagnosis based on retinal colour images. We focus on capturing the inherent colour changes of optic disc (OD) and cup borders by computing several colour derivatives in CIE L∗a∗b∗ colour space with CIE94 colour distance. In addition, we consider spatial information retaining these colour derivatives and the original CIE L∗a∗b∗ values of the pixel and adding other characteristics such as its distance to the OD centre. The proposed strategy is robust due to a simple structure that does not need neither initial segmentation nor removal of the vascular tree or detection of vessel bends. The method has been extensively validated with two datasets (one public and one private), each one comprising 60 images of high variability of appearances. Achieved class-wise-averaged accuracy of 95.02% and 81.19% demonstrates that this automated approach could support physicians in the diagnosis of glaucoma in its early stage, and therefore, it could be seen as an opportunity for developing low-cost solutions for mass screening programs.

Detección automatizada de microaneurismas mediante crecimiento de regiones y red neuronal Fuzzy Artmap / Automated detection of microaneurysms by using region growing and fuzzy artmap neural network

Objetivo: Comprobar si las modificaciones metodológicas de este nuevo algoritmo mejoran el resultado de otra estrategia presentada anteriormente. Métodos: Se realza y filtra la imagen negada del canal verde de la retinografía digital en color. Se aplica una umbralización multitolerancia para obtener puntos candidatos y en cada semilla se realiza un crecimiento de regiones por variación de intensidades. Se toman 15 características de cada región y entrenamos una red neuronal Fuzzy Artmap con 42 retinografías. Se aplica la red en el estudio de 11 retinografías del programa de detección precoz de retinopatía diabética, de buena calidad, con lesiones iniciales, obtenidas con el retinógrafo no midriático Topcon NW200. Resultados: Dos oftalmólogos experimentados detectan 52 microaneurismas en las 11 imágenes. El algoritmo detecta 39 microaneurismas y 3.752 regiones más, confirmando 38 microaneurismas y 135 falsos positivos. La sensibilidad ha mejorado respecto al algoritmo anterior del 60,53 al 73,08%. Los falsos positivos has disminuido de 41,8 por imagen a 12,27. Conclusiones: El nuevo algoritmo presenta indudables mejoras respecto al anterior, pero aún se puede perfeccionar, sobre todo en la determinación inicial de semillas(AU)

Objective: To assess whether the methodological changes of this new algorithm improves the results of a previously presented strategy. Methods: We enhance the image and filter out the green channel of the digital color retinography. Multitolerance thresholding was applied to obtain candidate points and make a seed growing region by varying intensities. We took 15 characteristics from each region to train a Fuzzy Artmap neural network using 42 retinal photographs. This network was then applied in the study of 11 good quality retinal photographs included in the diabetic retinopathy early detection screening program, with initial stages of retinopathy, obtained with the Topcon NW200 non-mydriatic retinal camera. Results: Two experienced ophthalmologists detected 52 microaneurysms in 11 images. The algorithm detected 39 microaneurysms and 3,752 more regions, confirming 38 microaneurysm and 135 false positives. The sensitivity is improved compared to the previous algorithm, from 60.53 to 73.08%. False positives have dropped from 41.8 to 12.27 per image. Conclusions: The new algorithm is better than the previous one, but there is still room for improvement, especially in the initial determination of seeds(AU)

[Automated detection of microaneurysms by using region growing and Fuzzy Artmap neural network]. / Detección automatizada de microaneurismas mediante crecimiento de regiones y red neuronal Fuzzy Artmap.

OBJECTIVE: To assess whether the methodological changes of this new algorithm improves the results of a previously presented strategy. METHODS: We enhance the image and filter out the green channel of the digital color retinography. Multitolerance thresholding was applied to obtain candidate points and make a seed growing region by varying intensities. We took 15 characteristics from each region to train a Fuzzy Artmap neural network using 42 retinal photographs. This network was then applied in the study of 11 good quality retinal photographs included in the diabetic retinopathy early detection screening program, with initial stages of retinopathy, obtained with the Topcon NW200 non-mydriatic retinal camera. RESULTS: Two experienced ophthalmologists detected 52 microaneurysms in 11 images. The algorithm detected 39 microaneurysms and 3,752 more regions, confirming 38 microaneurysm and 135 false positives. The sensitivity is improved compared to the previous algorithm, from 60.53 to 73.08%. False positives have dropped from 41.8 to 12.27 per image. CONCLUSIONS: The new algorithm is better than the previous one, but there is still room for improvement, especially in the initial determination of seeds.

[Automatic detection of microaneurysms in colour fundus images]. / Detección automática de microaneurismas en retinografías.

PURPOSE: We present the development of a tool for the automatic detection of microaneurysms and its clinical evaluation. The intention of this tool is to facilitate the diagnosis of diabetic retinopathy in general screening programs. METHOD: The designed and developed tool consists of three stages of processing: 1) Obtaining of the basic image of eye with the retinal camera, inverted image on the green channel, and a high-pass filter of the image. This phase enhances the microaneurysms. 2) Detection of the candidates for microaneurysms, by means of an adaptive prediction filter and regions growth. 3) Selection, among the candidates, of whom microaneurysms must be considered to fulfil the criteria of circular shape, high intensity in the inverted green channel and contrasts with respect to the surrounding pixels. RESULTS: We selected to 20 retinal photographs of good quality and dimensions 600x600 pixels from patients with nonproliferative diabetic retinopathy. The ophthalmologists detected 297 microaneurysms in these images. The tool for automatic detection correctly located 252 microaneurysms, with a mean sensitivity of 89% and a false positives rate of 93%. CONCLUSIONS: The results obtained seem to indicate that the tool developed will be very useful for its potential use in screening programs in primary care centres. On the other hand, more work is needed on the algorithm to decrease the rate of false positives.

Detección automática de microaneurismas en retinografías / Automatic detection of microaneurysms in colour fundus images

Propósito: Presentamos el desarrollo de una herramienta para la detección automática de microaneurismas y su evaluación clínica. El propósito de esta herramienta es facilitar el diagnóstico de lesiones diabéticas en programas generales de detección.MétodoLa herramienta diseñada y desarrollada consta de tres etapas de procesamiento: 1) Obtención de la imagen de fondo de ojo con el retinógrafo, inversión del canal verde y filtrado paso de alta de la imagen. Esta fase realza los microaneurismas. 2) Detección de los candidatos a microaneurismas, mediante un filtrado de predicción adaptativo y un crecimiento de regiones. 3) Selección, de entre los candidatos, de los que deben considerarse microaneurismas por cumplir con los criterios de: forma circular, intensidad alta en el canal verde invertido y contraste respecto a los píxeles de alrededor.ResultadosSe seleccionaron 20 retinografías de buena calidad y dimensiones 600×600 píxeles de pacientes con retinopatía diabética no proliferante. Los oftalmólogos detectaron un total de 297 microaneurismas en estas imágenes. La herramienta de detección automática localizó adecuadamente 252 microaneurismas, con una sensibilidad media del 89% y una tasa de falsos positivos del 93%.ConclusionesLos resultados obtenidos parecen indicar que la herramienta desarrollada podría ser muy útil para su potencial utilización en programas de detección en los centros de asistencia primaria. Por otro lado, es necesario seguir trabajando en el algoritmo para disminuir la tasa de falsos positivos (AU)

Purpose: We present the development of a tool for the automatic detection of microaneurysms and its clinical evaluation. The intention of this tool is to facilitate the diagnosis of diabetic retinopathy in general screening programs.MethodThe designed and developed tool consists of three stages of processing: 1) Obtaining of the basic image of eye with the retinal camera, inverted image on the green channel, and a high-pass filter of the image. This phase enhances the microaneurysms. 2) Detection of the candidates for microaneurysms, by means of an adaptive prediction filter and regions growth. 3) Selection, among the candidates, of whom microaneurysms must be considered to fulfil the criteria of circular shape, high intensity in the inverted green channel and contrasts with respect to the surrounding pixels.ResultsWe selected to 20 retinal photographs of good quality and dimensions 600x600 pixels from patients with nonproliferative diabetic retinopathy. The ophthalmologists detected 297 microaneurysms in these images. The tool for automatic detection correctly located 252 microaneurysms, with a mean sensitivity of 89% and a false positives rate of 93%.ConclusionsThe results obtained seem to indicate that the tool developed will be very useful for its potential use in screening programs in primary care centres. On the other hand, more work is needed on the algorithm to decrease the rate of false positives (AU)

[Automatic detection of vessels in color fundus images]. / Detección automática de vasos en retinografías.

PURPOSE: The main purpose of the paper is to evaluate an automated method for blood vessels segmentation in color fundus images, due to its important role in the diagnosis of several pathologies such as diabetes. The final objective is to introduce the algorithm into a Computer Aided Diagnosis (CAD) tool that would be available in those local medical centers without specialists. METHOD: An automated method for blood vessels segmentation in color fundus images was implemented and tested. The algorithm starts with the extraction of vessel centerlines, which are used as guidelines for the subsequent vessel filling phase. The outputs of four directional differential operators are processed in order to select connected sets of candidate points to be further classified as centerline pixels using vessel derived features. The final segmentation is obtained using an iterative region growing method that integrates the contents of several binary images, resulting from vessel width dependent morphological filters. The method was evaluated using the images of two publicly available databases (STARE and DRIVE) and a database with 24 images. RESULTS: The algorithm outperforms other published algorithms and approximates the average accuracy of a human observer without a significant degradation of sensitivity and specificity. In addition, results have been subject to the experts' valuation that they think that retinal vessels remain represented with valuable accuracy on having analyzed the test's images. CONCLUSION: Due to the good segmentation results, the algorithm proposed could be implemented as part of a complete CAD tool in the local medical centers. This would reduce cost and diagnosis time.

Detección automática de vasos en retinografías / Automatic detection of vessels in color fundus images

Propósito: El propósito de este trabajo es la evaluación de un método automático para lasegmentación del árbol vascular en imágenes de retinografías, dado su importante papelen el diagnóstico de numerosas enfermedades, como la diabetes mellitus. El objetivo finales introducir el algoritmo en una herramienta de diagnóstico asistido por computadora(CAD, del inglés Computer Aided Diagnosis) que estaría disponible en los centros médicoslocales sin especialistas.Método: Se ha implementado y probado un método automático para la segmentación devasos. El algoritmo comienza con la extracción de las líneas centrales de los vasos, que seemplean como guías para la fase posterior de rellenado de vasos. Las salidas de 4 operadoresdireccionales se procesan para obtener conjuntos conexos de puntos candidatos que seclasificarán como píxeles pertenecientes a las líneas centrales mediante característicasderivadas de los vasos. La segmentación final se obtiene empleando un proceso iterativode crecimiento de regiones que integra los contenidos de varias imágenes binarias, resultadode aplicar determinados filtros morfológicos que dependen del ancho del vaso. Elmétodo se ha evaluado empleando las imágenes de 2 bases de datos públicas (STARE yDRIVE) y por una base de datos compuesta por 24 imágenes.Resultados: El algoritmo mejora otras soluciones y se aproxima en precisión a la obtenidapor un observador humano, sin por ello experimentar una degradación de la sensibilidady la especificidad. Asimismo, los resultados del algoritmo se han sometido a la valoraciónde expertos que consideran que los vasos quedan representados con apreciable exactitudal analizar las imágenes de prueba.Conclusión: Dados los buenos resultados obtenidos en la segmentación, el algoritmo propuestopodría implementarse e introducirse en una herramienta CAD disponible en los centrosmédicos locales. La reducción en coste y tiempo de exploración podría ser significativa(AU)

Purpose: The main purpose of the paper is to evaluate an automated method for bloodvessels segmentation in color fundus images, due to its important role in the diagnosis ofseveral pathologies such as diabetes. The final objective is to introduce the algorithm intoa Computer Aided Diagnosis (CAD) tool that would be available in those local medicalcenters without specialists.Method: An automated method for blood vessels segmentation in color fundus images wasimplemented and tested. The algorithm starts with the extraction of vessel centerlines,which are used as guidelines for the subsequent vessel filling phase. The outputs of fourdirectional differential operators are processed in order to select connected sets ofcandidate points to be further classified as centerline pixels using vessel derived features.The final segmentation is obtained using an iterative region growing method thatintegrates the contents of several binary images, resulting from vessel width dependentmorphological filters. The method was evaluated using the images of two publicly availabledatabases (STARE and DRIVE) and a database with 24 images.Results: The algorithm outperforms other published algorithms and approximates theaverage accuracy of a human observer without a significant degradation of sensitivity andspecificity. In addition, results have been subject to the experts’ valuation that they thinkthat retinal vessels remain represented with valuable accuracy on having analyzed thetest’s images.Conclusion: Due to the good segmentation results, the algorithm proposed could beimplemented as part of a complete CAD tool in the local medical centers. This wouldreduce cost and diagnosis time(AU)