Attila Budai

Retinal vessel segmentation by improved matched filtering: evaluation on a new high-resolution fundus image database

Automatic assessment of retinal vessels plays an important role in the diagnosis of various eye, as well as systemic diseases. A public screening is highly desirable for prompt and effective treatment, since such diseases need to be diagnosed at an early stage. Automated and accurate segmentation of the retinal blood vessel tree is one of the challenging tasks in the computer-aided analysis of fundus images today. We improve the concept of matched filtering, and propose a novel and accurate method for segmenting retinal vessels. Our goal is to be able to segment blood vessels with varying vessel diameters in high-resolution colour fundus images. All recent authors compare their vessel segmentation results to each other using only low-resolution retinal image databases. Consequently, we provide a new publicly available high-resolution fundus image database of healthy and pathological retinas. Our performance evaluation shows that the proposed blood vessel segmentation approach is at least comparable with recent state-of-the-art methods. It outperforms most of them with an accuracy of 95% evaluated on the new database.

Quantitative 3D-OCT motion correction with tilt and illumination correction, robust similarity measure and regularization

Variability in illumination, signal quality, tilt and the amount of motion pose challenges for post-processing based 3D-OCT motion correction algorithms. We present an advanced 3D-OCT motion correction algorithm using image registration and orthogonal raster scan patterns aimed at addressing these challenges. An intensity similarity measure using the pseudo Huber norm and a regularization scheme based on a pseudo L0.5 norm are introduced. A two-stage registration approach was developed. In the first stage, only axial motion and axial tilt are coarsely corrected. This result is then used as the starting point for a second stage full optimization. In preprocessing, a bias field estimation based approach to correct illumination differences in the input volumes is employed. Quantitative evaluation was performed using a large set of data acquired from 73 healthy and glaucomatous eyes using SD-OCT systems. OCT volumes of both the optic nerve head and the macula region acquired with three independent orthogonal volume pairs for each location were used to assess reproducibility. The advanced motion correction algorithm using the techniques presented in this paper was compared to a basic algorithm corresponding to an earlier version and to performing no motion correction. Errors in segmentation-based measures such as layer positions, retinal and nerve fiber thickness, as well as the blood vessel pattern were evaluated. The quantitative results consistently show that reproducibility is improved considerably by using the advanced algorithm, which also significantly outperforms the basic algorithm. The mean of the mean absolute retinal thickness difference over all data was 9.9 um without motion correction, 7.1 um using the basic algorithm and 5.0 um using the advanced algorithm. Similarly, the blood vessel likelihood map error is reduced to 69% of the uncorrected error for the basic and to 47% of the uncorrected error for the advanced algorithm. These results demonstrate that our advanced motion correction algorithm has the potential to improve the reliability of quantitative measurements derived from 3D-OCT data substantially.

Robust Vessel Segmentation in Fundus Images

One of the most common modalities to examine the human eye is the eye-fundus photograph. The evaluation of fundus photographs is carried out by medical experts during time-consuming visual inspection. Our aim is to accelerate this process using computer aided diagnosis. As a first step, it is necessary to segment structures in the images for tissue differentiation. As the eye is the only organ, where the vasculature can be imaged in an in vivo and noninterventional way without using expensive scanners, the vessel tree is one of the most interesting and important structures to analyze. The quality and resolution of fundus images are rapidly increasing. Thus, segmentation methods need to be adapted to the new challenges of high resolutions. In this paper, we present a method to reduce calculation time, achieve high accuracy, and increase sensitivity compared to the original Frangi method. This method contains approaches to avoid potential problems like specular reflexes of thick vessels. The proposed method is evaluated using the STARE and DRIVE databases and we propose a new high resolution fundus database to compare it to the state-of-the-art algorithms. The results show an average accuracy above 94% and low computational needs. This outperforms state-of-the-art methods.

Automatic no-reference quality assessment for retinal fundus images using vessel segmentation

Fundus imaging is the most commonly used modality to collect information about the human eye background. Objective and quantitative assessment of quality for the acquired images is essential for manual, computer-aided and fully automatic diagnosis. In this paper, we present a no-reference quality metric to quantify image noise and blur and its application to fundus image quality assessment. The proposed metric takes the vessel tree visible on the retina as guidance to determine an image quality score. In our experiments, the performance of this approach is demonstrated by correlation analysis with the established full-reference metrics peak-signal-to-noise ratio (PSNR) and structural similarity (SSIM). We found a Spearman rank correlation for PSNR and SSIM of 0.89 and 0.91. For real data, our metric correlates reasonable to a human observer, indicating high agreement to human visual perception.

Optic disk localization using fast radial symmetry transform

Fundus imaging is one of the most frequently used modalities for screening, diagnosis of eye diseases and some vascular abnormalities. Due to its wide availability, automatic evaluation of fundus images offers great potential benefits to current clinical practice. The basis of many automatic evaluations or diagnosis is the segmentation of the eye background, most notably, the detection of the optic disk and the segmentation of the retinal vessel tree. In this work we propose a variant of the fast radial symmetry transform (FRST), adapted to its application in the detection of the optic disk in fundus images. We evaluated and compared the performance of our method to the standard FRST and the similar, gradient based circular Hough transform using 45 images of a high resolution database with gold standard information available. We demonstrated in our experiments that the proposed method outperforms the state-of-the-art algorithms with 0.051 ± 0.073 optic disk diameter localization error in average.

Geometry-Based Optic Disk Tracking in Retinal Fundus Videos

Fundus video cameras enable the acquisition of image se- quences to analyze fast temporal changes on the human retina in a non-invasive manner. In this work, we propose a tracking-by-detection scheme for the optic disk to capture the human eye motion on-line dur- ing examination. Our approach exploits the elliptical shape of the optic disk. Therefore, we employ the fast radial symmetry transform for an ef- ficient estimation of the disk center point in successive frames. Large eye movements due to saccades, motion of the head or blinking are detected automatically by a correlation analysis to guide the tracking procedure. In our experiments on real video data acquired by a low-cost video cam- era, the proposed method yields a hit rate of 98% with a normalized median accuracy of 4% of the disk diameter. The achieved frame rate of 28 frames per second enables a real-time application of our approach.

The Role of Retinal Vascular Density as a Screening Tool for Ageing and Stroke

Objectives: To measure the density of retinal vessels from digitized fundus photographs in patients with recent stroke and age-matched controls. To investigate whether the parameter retinal vascular density (RVD) served as a quantitative marker for cerebrovascular events. Methods: Digitized fundus photographs of n = 158 subjects with stroke or transient ischemic attack within 1 year at the time of examination and n = 1,250 age-matched controls without any remarkable medical history were examined. Sex, hypertension, and diabetes were considered to be cofactors. Measurement of RVD was performed with a computer-aided image-analyzing program by segmenting automatically all visible retinal vessels and measuring areas of vessels in distinct circles around the optic disk. Results: In controls RVD dwindles with increasing distance from the optic disk. RVD decreased significantly with age (p = 0.000). Stroke patients showed significantly lower values of RVD of –15% in comparison to age-matched controls. In old subjects, stroke in combination with hypertension is associated with a significant decreased RVD, and in middle-aged subjects diabetes and stroke are associated with a significant decreased RVD (p = 0.01). Conclusion: Age and stroke are significant risk factors for decreased RVD. Diabetes and arterial hypertension are additional significant risk factors in patients with stroke with respect to RVD.

Analysis of the Retinal Nerve Fiber Layer Texture Related to the Thickness Measured by Optical Coherence Tomography

The retinal nerve fiber layer (RNFL) is one of the most affected retinal structures due to the glaucoma disease. Progression of this disease results in the RNFL atrophy that can be detected as the decrease of the layer’s thickness. Usually, the RNFL thickness can be assessed by optical coherence tomography (OCT). However, an examination using OCT is rather expensive and still not widely available. On the other hand, fundus camera is considered as a common and fundamental diagnostic device utilized at many ophthalmic facilities worldwide. This contribution presents a novel approach to texture analysis enabling assessment of the RNFL thickness in widely used colour fundus photographs. The aim is to propose a regression model based on different texture features effective for description of changes in the RNFL textural appearance related to the variations of RNFL thickness. The performance evaluation uses OCT as a gold standard modality for validation of the proposed approach. The results show high correlation between the models predicted output and RNFL thickness directly measured by OCT.

Blood vessel segmentation in video-sequences from the human retina

This paper deals with the retinal blood vessel segmentation in fundus video-sequences acquired by experimental fundus video camera. Quality of acquired video-sequences is relatively low and fluctuates across particular frames. Especially, due to the low resolution, poor signal-to-noise ratio, and varying illumination conditions within the frames, application of standard image processing methods might be difficult in such experimental fundus images. In this study, we tried two methods for the segmentation of retinal vessels - matched filtering and Hessian-based approach, originally developed for vessel segmentation in standard fundus images. We showed that modified versions of these two approaches, combined with support vector machine (SVM), can be used also for segmentation in experimental low-quality fundus video-sequences. The SVM classifier trained and consecutively tested on the database of high-resolution images achieved classification accuracy over 94 % and thus revealed a possible applicability of the proposed method on low-quality data. Then, testing on low-quality video-sequences revealed sufficiently large reliability in term of segmentation stability within the sequence with the interframe variability in image quality.

Automatic Fovea Localization in Fundus Images

One of the most common modalities to examine the eye- background is the fundus image. These images are photographs taken through the pupil by a fundus camera. The evaluation of these im- ages is usually carried out by ophthalmologists or experts during visual inspection. The aim of our work is to accelerate this process by a fully- automatic screening. To enable the automatic disease and tissue specific feature extraction, it is necessary to segment the different visible struc- tures in the images. Our group already published methods to extract the vessel tree and estimate the position and diameter of the optic nerve head. In this paper, we present our methods to localize the fovea and the macula region, and compare it to other approaches using the High Resolution Fundus database. Our evaluation showed, that our proposed method is capable of localizing the macula region with an average dis- tance error of 0.1 times optic disk diameter.