Radim Kolar

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.

Retinal image analysis aimed at blood vessel tree segmentation and early detection of neural-layer deterioration

An automatic method of segmenting the retinal vessel tree and estimating status of retinal neural fibre layer (NFL) from high resolution fundus camera images is presented. First, reliable blood vessel segmentation, using 2D directional matched filtering, enables to remove areas occluded by blood vessels thus leaving remaining retinal area available to the following NFL detection. The local existence of rather faint and hardly visible NFL is detected by combining several newly designed local textural features, sensitive to subtle NFL characteristics, into feature vectors submitted to a trained neural-network classifier. Obtained binary retinal maps of NFL distribution show a good agreement with both medical expert evaluations and quantitative results obtained by optical coherence tomography.

Hybrid retinal image registration using phase correlation

Abstract This paper deals with registration of retinal images, which were taken by high-resolution digital colour fundus cameras. The proposed method describes successful application of phase correlation method. It combines several basic steps — global correction of shift, rotation and scaling, detection of landmarks, their correspondences and finally image registration using second-order polynomial model. The method is tested on two sets of images. The first set contains images from the diabetic patients where many retinal pathologies can disturb the registration process. The second set contains images from healthy subjects, which were acquired by different illumination conditions. The method was evaluated using four different criteria - tree objective and one subjective. These criteria are also compared. The achieved registration accuracy of the landmarks position error is 1·13 and 0·93 pixels for respective image sets. Finally, the simple scheme for retinal pathology visualisation of registered fundus pairs is presented.

Ultrasound perfusion analysis combining bolus-tracking and burst-replenishment

A new signal model and processing method for quantitative ultrasound perfusion analysis is presented, called bolus-and-burst. The method has the potential to provide absolute values of blood flow, blood volume, and mean transit time. Furthermore, it provides an estimate of the local arterial input function which characterizes the arterial tree, allowing accurate estimation of the bolus arrival time. The method combines two approaches to ultrasound perfusion analysis: bolus-tracking and burst-replenishment. A pharmacokinetic model based on the concept of arterial input functions and tissue residue functions is used to model both the bolus and replenishment parts of the recording. The pharmacokinetic model is fitted to the data using blind deconvolution. A preliminary assessment of the new perfusion-analysis method is presented on clinical recordings.

Retrospective illumination correction of retinal images

A method for correction of nonhomogenous illumination based on optimization of parameters of B-spline shading model with respect to Shannons entropy is presented. The evaluation of Shannons entropy is based on Parzen windowing method (Mangin, 2000) with the spline-based shading model. This allows us to express the derivatives of the entropy criterion analytically, which enables efficient use of gradient-based optimization algorithms. Seven different gradient- and nongradient-based optimization algorithms were initially tested on a set of 40 simulated retinal images, generated by a model of the respective image acquisition system. Among the tested optimizers, the gradient-based optimizer with varying step has shown to have the fastest convergence while providing the best precision. The final algorithm proved to be able of suppressing approximately 70% of the artificially introduced non-homogenous illumination. To assess the practical utility of the method, it was qualitatively tested on a set of 336 real retinal images; it proved the ability of eliminating the illumination inhomogeneity substantially in most of cases. The application field of this method is especially in preprocessing of retinal images, as preparation for reliable segmentation or registration.

Registration and Fusion of the Autofluorescent and Infrared Retinal Images

This article deals with registration and fusion of multimodal opththalmologic images obtained by means of a laser scanning device (Heidelberg retina angiograph). The registration framework has been designed and tested for combination of autofluorescent and infrared images. This process is a necessary step for consecutive pixel level fusion and analysis utilizing information from both modalities. Two fusion methods are presented and compared.

Parametric ultrasound perfusion analysis combining bolus tracking and replenishment

The paper presents a new perfusion analysis method using ultrasound which combines burst-replenishment and bolustracking acquisition methods. It allows absolute quantification of the mean transit time, blood flow and blood volume. It is based on the concept of arterial input function and tissue residue function and is formulated as a blind-deconvolution problem. It is illustrated on recordings from Crohns disease patients.

Analysis of Visual Appearance of Retinal Nerve Fibers in High Resolution Fundus Images: A Study on Normal Subjects

The retinal ganglion axons are an important part of the visual system, which can be directly observed by fundus camera. The layer they form together inside the retina is the retinal nerve fiber layer (RNFL). This paper describes results of a texture RNFL analysis in color fundus photographs and compares these results with quantitative measurement of RNFL thickness obtained from optical coherence tomography on normal subjects. It is shown that local mean value, standard deviation, and Shannon entropy extracted from the green and blue channel of fundus images are correlated with corresponding RNFL thickness. The linear correlation coefficients achieved values 0.694, 0.547, and 0.512 for respective features measured on 439 retinal positions in the peripapillary area from 23 eyes of 15 different normal subjects.

Comparison and evaluation of indicator dilution models for bolus of ultrasound contrast agents.

Dynamic contrast-enhanced ultrasound (DCE-US) imaging is a promising diagnostic method, which enables the evaluation of tissue perfusion via different parameters. The mean transit time and time-to-peak parameters are the main time parameters and their values depend on the model used for the approximation of the noisy perfusion curves. In this paper, we described a new comparison of different perfusion models using a tissue mimicking phantom. The following models were compared: log-normal, lagged, Erlang, Gamma and the local density random walk model. We discovered that the mean-square error is not the best criterion for model evaluation. More important is the comparison between the estimated time perfusion parameters and the physical parameters of the developed tissue mimicking phantom. Based on the statistical analysis, we can suggest that for the DCE-US perfusion analysis more models should be used, excluding the log-normal model, which gives the highest error of mean transit time value.

Registration of Image Sequences from Experimental Low-Cost Fundus Camera

This paper describes new registration approach for registration of low SNR retinal image sequences. We combine two approaches - Fourier-based method for large shift correction and Lucas-Kanade tracking for small shift and rotation correction. We also propose method for evaluation of registration results, which uses spatial variation of minimum value in intensity profiles through blood-vessels. We achieved precision of registration below 2.1 pixels, which is acceptable with regards to image SNR (around 10dB). The final averaging of registered sequence leads to improvement of image quality and improvement in SNR over 10 dB.