Michael D. Abràmoff

Ridge-based vessel segmentation in color images of the retina

A method is presented for automated segmentation of vessels in two-dimensional color images of the retina. This method can be used in computer analyses of retinal images, e.g., in automated screening for diabetic retinopathy. The system is based on extraction of image ridges, which coincide approximately with vessel centerlines. The ridges are used to compose primitives in the form of line elements. With the line elements an image is partitioned into patches by assigning each image pixel to the closest line element. Every line element constitutes a local coordinate frame for its corresponding patch. For every pixel, feature vectors are computed that make use of properties of the patches and the line elements. The feature vectors are classified using a kNN-classifier and sequential forward feature selection. The algorithm was tested on a database consisting of 40 manually labeled images. The method achieves an area under the receiver operating characteristic curve of 0.952. The method is compared with two recently published rule-based methods of Hoover et al. and Jiang et al. . The results show that our method is significantly better than the two rule-based methods (p<0.01). The accuracy of our method is 0.944 versus 0.947 for a second observer.

Comparative study of retinal vessel segmentation methods on a new publicly available database

In this work we compare the performance of a number of vessel segmentation algorithms on a newly constructed retinal vessel image database. Retinal vessel segmentation is important for the detection of numerous eye diseases and plays an important role in automatic retinal disease screening systems. A large number of methods for retinal vessel segmentation have been published, yet an evaluation of these methods on a common database of screening images has not been performed. To compare the performance of retinal vessel segmentation methods we have constructed a large database of retinal images. The database contains forty images in which the vessel trees have been manually segmented. For twenty of those forty images a second independent manual segmentation is available. This allows for a comparison between the performance of automatic methods and the performance of a human observer. The database is available to the research community. Interested researchers are encouraged to upload their segmentation results to our website (http://www.isi.uu.nl/Research/Databases). The performance of five different algorithms has been compared. Four of these methods have been implemented as described in the literature. The fifth pixel classification based method was developed specifically for the segmentation of retinal vessels and is the only supervised method in this test. We define the segmentation accuracy with respect to our gold standard as the performance measure. Results show that the pixel classification method performs best, but the second observer still performs significantly better.

Automatic detection of red lesions in digital color fundus photographs

The robust detection of red lesions in digital color fundus photographs is a critical step in the development of automated screening systems for diabetic retinopathy. In this paper, a novel red lesion detection method is presented based on a hybrid approach, combining prior works by Spencer et al. (1996) and Frame et al. (1998) with two important new contributions. The first contribution is a new red lesion candidate detection system based on pixel classification. Using this technique, vasculature and red lesions are separated from the background of the image. After removal of the connected vasculature the remaining objects are considered possible red lesions. Second, an extensive number of new features are added to those proposed by Spencer-Frame. The detected candidate objects are classified using all features and a k-nearest neighbor classifier. An extensive evaluation was performed on a test set composed of images representative of those normally found in a screening set. When determining whether an image contains red lesions the system achieves a sensitivity of 100% at a specificity of 87%. The method is compared with several different automatic systems and is shown to outperform them all. Performance is close to that of a human expert examining the images for the presence of red lesions.

Automated 3-D Intraretinal Layer Segmentation of Macular Spectral-Domain Optical Coherence Tomography Images

With the introduction of spectral-domain optical coherence tomography (OCT), much larger image datasets are routinely acquired compared to what was possible using the previous generation of time-domain OCT. Thus, the need for 3-D segmentation methods for processing such data is becoming increasingly important. We report a graph-theoretic segmentation method for the simultaneous segmentation of multiple 3-D surfaces that is guaranteed to be optimal with respect to the cost function and that is directly applicable to the segmentation of 3-D spectral OCT image data. We present two extensions to the general layered graph segmentation method: the ability to incorporate varying feasibility constraints and the ability to incorporate true regional information. Appropriate feasibility constraints and cost functions were learned from a training set of 13 spectral-domain OCT images from 13 subjects. After training, our approach was tested on a test set of 28 images from 14 subjects. An overall mean unsigned border positioning error of 5.69 plusmn 2.41 mum was achieved when segmenting seven surfaces (six layers) and using the average of the manual tracings of two ophthalmologists as the reference standard. This result is very comparable to the measured interobserver variability of 5.71 plusmn 1.98 mum.

Retinopathy Online Challenge: Automatic Detection of Microaneurysms in Digital Color Fundus Photographs

The detection of microaneurysms in digital color fundus photographs is a critical first step in automated screening for diabetic retinopathy (DR), a common complication of diabetes. To accomplish this detection numerous methods have been published in the past but none of these was compared with each other on the same data. In this work we present the results of the first international microaneurysm detection competition, organized in the context of the Retinopathy Online Challenge (ROC), a multiyear online competition for various aspects of DR detection. For this competition, we compare the results of five different methods, produced by five different teams of researchers on the same set of data. The evaluation was performed in a uniform manner using an algorithm presented in this work. The set of data used for the competition consisted of 50 training images with available reference standard and 50 test images where the reference standard was withheld by the organizers (M. Niemeijer, B. van Ginneken, and M. D. AbrA¿moff). The results obtained on the test data was submitted through a website after which standardized evaluation software was used to determine the performance of each of the methods. A human expert detected microaneurysms in the test set to allow comparison with the performance of the automatic methods. The overall results show that microaneurysm detection is a challenging task for both the automatic methods as well as the human expert. There is room for improvement as the best performing system does not reach the performance of the human expert. The data associated with the ROC microaneurysm detection competition will remain publicly available and the website will continue accepting submissions.

Selective Loss of Inner Retinal Layer Thickness in Type 1 Diabetic Patients with Minimal Diabetic Retinopathy

PURPOSE To determine whether type 1 diabetes preferentially affects the inner retinal layers by comparing the thickness of six retinal layers in type 1 diabetic patients who have no or minimal diabetic retinopathy (DR) with those of age- and sex-matched healthy controls. METHODS Fifty-seven patients with type 1 diabetes with no (n = 32) or minimal (n = 25) DR underwent full ophthalmic examination, stereoscopic fundus photography, and optical coherence tomography (OCT). After automated segmentation of intraretinal layers of the OCT images, mean thickness was calculated for six layers of the retina in the fovea, the pericentral area, and the peripheral area of the central macula and were compared with those of an age- and sex-matched control group. RESULTS In patients with minimal DR, the mean ganglion cell/inner plexiform layer was 2.7 microm thinner (95% confidence interval [CI], 2.1-4.3 microm) and the mean inner nuclear layer was 1.1 microm thinner (95% CI, 0.1-2.1 microm) in the pericentral area of the central macula compared to those of age-matched controls. In the peripheral area, the mean ganglion cell/inner plexiform layer remained significantly thinner. No other layers showed a significant difference. CONCLUSIONS Thinning of the total retina in type 1 diabetic patients with minimal retinopathy compared with healthy controls is attributed to a selective thinning of inner retinal layers and supports the concept that early DR includes a neurodegenerative component.

Decreased Retinal Ganglion Cell Layer Thickness in Patients with Type 1 Diabetes

PURPOSE. To determine which retinal layers are most affected by diabetes and contribute to thinning of the inner retina and to investigate the relationship between retinal layer thickness (LT) and diabetes duration, diabetic retinopathy (DR) status, age, glycosylated hemoglobin (HbA1c), and the sex of the individual, in patients with type 1 diabetes who have no or minimal DR. METHODS. Mean LT was calculated for the individual retinal layers after automated segmentation of spectral domain-optical coherence tomography scans of patients with diabetes and compared with that in control subjects. Multiple linear regression analysis was used to determine the relationship between LT and HbA1c, age, sex, diabetes duration, and DR status. RESULTS. In patients with minimal DR, the mean ganglion cell layer (GCL) in the pericentral area was 5.1 mum thinner (95% confidence interval [CI], 1.1-9.1 mum), and in the peripheral macula, the mean retinal nerve fiber layer (RNFL) was 3.7 mum thinner (95% CI, 1.3-6.1 mum) than in the control subjects. There was a significant linear correlation (R = 0.53, P < 0.01) between GCL thickness and diabetes duration in the pooled group of patients. Multiple linear regression analysis (R = 0.62, P < 0.01) showed that DR status was the most important explanatory variable. CONCLUSIONS. This study demonstrates GCL thinning in the pericentral area and corresponding loss of RNFL thickness in the peripheral macula in patients with type 1 diabetes and no or minimal DR compared with control subjects. These results support the concept that diabetes has an early neurodegenerative effect on the retina, which occurs even though the vascular component of DR is minimal.

Segmentation of the Optic Disc, Macula and Vascular Arch in Fundus Photographs

An automatic system is presented to find the location of the major anatomical structures in color fundus photographs; the optic disc, the macula, and the vascular arch. These structures are found by fitting a single point-distribution-model to the image, that contains points on each structure. The method can handle optic disc and macula centered images of both the left and the right eye. The system uses a cost function, which is based on a combination of both global and local cues, to find the correct position of the model points. The global terms in the cost function are based on the orientation and width of the vascular pattern in the image. The local term is derived from the image structure around the points of the model. To optimize the fit of the point-distribution-model to an image, a sophisticated combination of optimization processes is proposed which combines optimization in the parameter space of the model and in the image space, where points are moved directly. Experimental results are presented demonstrating that our specific choices for the cost function components and optimization scheme are needed to obtain good results. The system was developed and trained on a set of 500 screening images, and tested on a completely independent set of 500 screening images. In addition to this the system was also tested on a separate set of 100 pathological images. In the screening set it was able to find the vascular arch in 93.2%, the macula in 94.4%, the optic disc location in 98.4% and whether it is dealing with a left or right eye in 100% of all tested cases. For the pathological images test set, this was 77.0%, 92.0%, 94.0%, and 100% respectively

Multiscale AM-FM Methods for Diabetic Retinopathy Lesion Detection

In this paper, we propose the use of multiscale amplitude-modulation-frequency-modulation (AM-FM) methods for discriminating between normal and pathological retinal images. The method presented in this paper is tested using standard images from the early treatment diabetic retinopathy study. We use 120 regions of 40 × 40 pixels containing four types of lesions commonly associated with diabetic retinopathy (DR) and two types of normal retinal regions that were manually selected by a trained analyst. The region types included microaneurysms, exudates, neovascularization on the retina, hemorrhages, normal retinal background, and normal vessels patterns. The cumulative distribution functions of the instantaneous amplitude, the instantaneous frequency magnitude, and the relative instantaneous frequency angle from multiple scales are used as texture feature vectors. We use distance metrics between the extracted feature vectors to measure interstructure similarity. Our results demonstrate a statistical differentiation of normal retinal structures and pathological lesions based on AM-FM features. We further demonstrate our AM-FM methodology by applying it to classification of retinal images from the MESSIDOR database. Overall, the proposed methodology shows significant capability for use in automatic DR screening.

Fast detection of the optic disc and fovea in color fundus photographs

A fully automated, fast method to detect the fovea and the optic disc in digital color photographs of the retina is presented. The method makes few assumptions about the location of both structures in the image. We define the problem of localizing structures in a retinal image as a regression problem. A kNN regressor is utilized to predict the distance in pixels in the image to the object of interest at any given location in the image based on a set of features measured at that location. The method combines cues measured directly in the image with cues derived from a segmentation of the retinal vasculature. A distance prediction is made for a limited number of image locations and the point with the lowest predicted distance to the optic disc is selected as the optic disc center. Based on this location the search area for the fovea is defined. The location with the lowest predicted distance to the fovea within the foveal search area is selected as the fovea location. The method is trained with 500 images for which the optic disc and fovea locations are known. An extensive evaluation was done on 500 images from a diabetic retinopathy screening program and 100 specially selected images containing gross abnormalities. The method found the optic disc in 99.4% and the fovea in 96.8% of regular screening images and for the images with abnormalities these numbers were 93.0% and 89.0% respectively.