Fundus Image Quality Research Articles

Multivariate Regression-Based Convolutional Neural Network Model for Fundus Image Quality Assessment

Objectively assessing the perceptual quality of an ocular fundus image is essential for the reliable diagnosis of various ocular diseases. A fair amount of work has been done in this field to date. However, the generalizability of the current work is limited, as the existing quality models were developed and evaluated with data-sets built with limited subjective inputs. This paper aims at addressing this limitation with the following two contributions. First, a new fundus image quality assessment (FIQuA) data-set is presented, containing 1500 fundus images with three classes of quality: Good, Fair, and Poor. Also, for each image, subjective scores (in the range [0-10]) were collected for six quality parameters, including structural and generic properties of the fundus images. Second, a new multivariate regression based convolutional neural network (CNN) model is proposed to predict the fundus image quality. The proposed model consists of two individually trained blocks. The first block consists of four pre-trained models, trained against the subjective scores for the six quality parameters, and aims at deriving the optimized features for classification. Next, the optimized features from each of the four models are ensembled together and transferred to the second block for final classification. The proposed model achieves a strong correlation with the subjective scores, with the values 0.941 , 0.954 , 0.853 , and 0.401 obtained for SROCC, LCC, KCC, and RMSE respectively. Its classification accuracy is 95.66% over the FIQuA data-set, and 98.96% and 88.43% respectively over the two publicly available data-sets DRIMDB and EyeQ.

Automated quality classification of colour fundus images based on a modified residual dense block network

Fundus image-assisted diagnosis has become an intuitive and standard diagnostic technique in recent years. In hospital treatment and telemedicine, experts conduct pathological analysis and treat patients with fundus diseases using retinal images that are captured by fundus cameras. The quality of fundus images is therefore crucial for doctors to provide timely and accurate diagnoses of the diseases. However, the different experiences of fundus photographers result in different image qualities, and some images are marked as unreadable by the diagnostician, seriously delaying the treatment of patients. To solve the quality classification problem, a modified residual dense block convolution neural network (MRDB-CNN) is proposed in this paper. The two categories of images, “good quality” and “poor quality”, are used as the training data set and testing data set to train the CNNs and verify the classification results. The experimental results show that, compared with other existing network structures, the network proposed in this paper can classify the “good-quality” and “poor-quality” fundus images more accurately with an accuracy rate of 99.90%. In addition, the tests for the “ambiguous-quality” category further demonstrate that our method can obtain more detailed features of the fundus images and give an objective score to the image quality. In short, the MRDB-CNN can accurately classify the quality of fundus images while avoiding the complex preprocessing of traditional algorithms, and it satisfies the needs of hospital treatment and telemedicine for the real-time quality judgement of fundus images.

Retinal image quality assessment based on FA-Net

Objective To propose a deep learning-based retinal image quality classification network, FA-Net, to make convolutional neural network (CNN) more suitable for image quality assessment in eye disease screening system. Methods The main network of FA-Net was composed of VGG-19.On this basis, attention mechanism was added to the CNN.By using transfer learning method in training, the weight of ImageNet was used to initialize the network.The attention net is based on foreground extraction by extracting the blood vessel and suspected regions of lesion and assigning higher weights to region of interest to enhance the learning of these important areas. Results Total of 2 894 fundus images were used for training FA-Net.FA-Net achieved 97.65% classification accuracy on a test set containing 2 170 fundus images, with the sensitivity and specificity of 0.978 and 0.960, respectively, and the area under curve(AUC) was 0.995. Conclusions Compared with other CNNs, the proposed FA-Net has better classification performance and can evaluate retinal fundus image quality more accurately and efficiently.The network takes into account the human visual system (HVS) and human attention mechanism.By adding attention module into the VGG-19 network structure, the classification results can be better interpreted as well as better classification performance. Key words: Retinal image quality assessment; Retinal image enhancement; Convolutional neural networks; Deep learning

Computer-aided diagnosis of cataract using deep transfer learning

Cataract is a leading eye disease across the world. If cataract is not diagnosed in earlier stage, then it may lead to blindness. Earlier detection is the best way to control the risk and to avoid painful surgery. Thus, this paper mainly focuses on cataract detection from fundus retinal images. A computer-aided automatic cataract detection method is proposed to detect various stages of the cataract such as normal, mild, moderate, and severe from the fundus images. The proposed method uses the pre-trained convolutional neural network (CNN) for the transfer learning to carry out automatic cataract classification. Then pre-trained CNN model is used for the feature extraction and the extracted features are then applied to a support vector machine (SVM) classifier. The fundus cataract images are collected from the various open access datasets and labelled into four stages with the help of ophthalmologic experts. The four stage classification accuracy obtained is 92.91%. Since, the image quality is important in CNN, an image quality selection module is incorporated to decide the quality of fundus image for diagnosis. The revaluation of results based on the quality of fundus images is also presented. Based on the results, the proposed method proved to be an efficient method that uses pre-trained CNN as transfer learning for the classification of the cataract.

Fundus image quality assessment: survey, challenges, and future scope

Various ocular diseases, such as cataract, diabetic retinopathy, and glaucoma have affected a large proportion of the population worldwide. In ophthalmology, fundus photography is used for the diagnosis of such retinal disorders. Nowadays, the set-up of fundus image acquisition has changed from a fixed position to portable devices, making acquisition more vulnerable to distortions. However, a trustworthy diagnosis solely relies upon the quality of the fundus image. In recent years, fundus image quality assessment (IQA) has drawn much attention from researchers. This study presents a detailed survey of the fundus IQA research. The survey covers a comprehensive discussion on the factors affecting the fundus image quality and the real-time distortions. The fundus IQA algorithms have been analysed on the basis of the methodologies used and divided into three classes, namely: (i) similarity-based, (ii) segmentation-based, and (iii) machine learning based. In addition, limitations of state of the art in this research field are also presented with the possible solutions. The objective of this study is to provide a detailed information about the fundus IQA research with its significance, present status, limitations, and future scope. To the best of the authors' knowledge, this is the first survey paper on the fundus IQA research.

Quality and content analysis of fundus images using deep learning

Automatic retinal image analysis has remained an important topic of research in the last ten years. Various algorithms and methods have been developed for analysing retinal images. The majority of these methods use public retinal image databases for performance evaluation without first examining the retinal image quality. Therefore, the performance metrics reported by these methods are inconsistent. In this article, we propose a deep learning-based approach to assess the quality of input retinal images. The method begins with a deep learning-based classification that identifies the image quality in terms of sharpness, illumination and homogeneity, followed by an unsupervised second stage that evaluates the field definition and content in the image. Using the inter-database cross-validation technique, our proposed method achieved overall sensitivity, specificity, positive predictive value, negative predictive value and accuracy of above 90% when tested on 7007 images collected from seven different public databases, including our own developed database—the UoA-DR database. Therefore, our proposed method is generalised and robust, making it more suitable than alternative methods for adoption in clinical practice.

PixelBNN: Augmenting the PixelCNN with Batch Normalization and the Presentation of a Fast Architecture for Retinal Vessel Segmentation.

Analysis of retinal fundus images is essential for eye-care physicians in the diagnosis, care and treatment of patients. Accurate fundus and/or retinal vessel maps give rise to longitudinal studies able to utilize multimedia image registration and disease/condition status measurements, as well as applications in surgery preparation and biometrics. The segmentation of retinal morphology has numerous applications in assessing ophthalmologic and cardiovascular disease pathologies. Computer-aided segmentation of the vasculature has proven to be a challenge, mainly due to inconsistencies such as noise and variations in hue and brightness that can greatly reduce the quality of fundus images. The goal of this work is to collate different key performance indicators (KPIs) and state-of-the-art methods applied to this task, frame computational efficiency–performance trade-offs under varying degrees of information loss using common datasets, and introduce PixelBNN, a highly efficient deep method for automating the segmentation of fundus morphologies. The model was trained, tested and cross tested on the DRIVE, STARE and CHASE_DB1 retinal vessel segmentation datasets. Performance was evaluated using G-mean, Mathews Correlation Coefficient and F1-score, with the main success measure being computation speed. The network was 8.5× faster than the current state-of-the-art at test time and performed comparatively well, considering a 5× to 19× reduction in information from resizing images during preprocessing.

Automated Fundus Image Quality Assessment in Retinopathy of Prematurity Using Deep Convolutional Neural Networks

Accurate image-based ophthalmic diagnosis relies on fundus image clarity. This has important implications for the quality of ophthalmic diagnoses and for emerging methods such as telemedicine and computer-based image analysis. The purpose of this study was to implement a deep convolutional neural network (CNN) for automated assessment of fundus image quality in retinopathy of prematurity (ROP). Experimental study. Retinal fundus images were collected from preterm infants during routine ROP screenings. Six thousand one hundred thirty-nine retinal fundus images were collected from 9 academic institutions. Each image was graded for quality (acceptable quality [AQ], possibly acceptable quality [PAQ], or not acceptable quality [NAQ]) by 3 independent experts. Quality was defined as the ability to assess an image confidently for the presence of ROP. Of the 6139 images, NAQ, PAQ, and AQ images represented 5.6%, 43.6%, and 50.8% of the image set, respectively. Because of low representation of NAQ images in the data set, images labeled NAQ were grouped into the PAQ category, and a binary CNN classifier was trained using 5-fold cross-validation on 4000 images. A test set of 2109 images was held out for final model evaluation. Additionally, 30 images were ranked from worst to best quality by 6 experts via pairwise comparisons, and the CNN's ability to rank quality, regardless of quality classification, was assessed. The CNN performance was evaluated using area under the receiver operating characteristic curve (AUC). A Spearman's rank correlation was calculated to evaluate the overall ability of the CNN to rank images from worst to best quality as compared with experts. The mean AUC for 5-fold cross-validation was 0.958 (standard deviation, 0.005) for the diagnosis of AQ versus PAQ images. The AUC was 0.965 for the test set. The Spearman's rank correlation coefficient on the set of 30 images was 0.90 as compared with the overall expert consensus ranking. This model accurately assessed retinal fundus image quality in a comparable manner with that of experts. This fully automated model has potential for application in clinical settings, telemedicine, and computer-based image analysis in ROP and for generalizability to other ophthalmic diseases.

Automated Quality Assessment of Fundus Images via Analysis of Illumination, Naturalness and Structure

In remote medical diagnosis, the percentage of poor-quality fundus images is very high, which requires automated quality assessment of fundus images in the acquisition stage to reduce the retransmission cost. In this paper, we propose a fundus image quality classifier via the analysis of illumination, naturalness, and structure, which use three effective secondary indices (or 5-D feature set) and different classification methods to determine the recommendation indexes of fundus images for further diagnosis. We construct a fundus image database including ‘accept’ and ‘reject’ classes based on the definition of illumination, naturalness, and structure. The model can achieve a sensitivity of 94.69%, specificity of 92.29%, and accuracy of 93.60% for the classifying of the fundus images.

Computer-aided diagnosis based on enhancement of degraded fundus photographs.

Retinal imaging is an important and effective tool for detecting retinal diseases. However, degraded images caused by the aberrations of the eye can disguise lesions, so that a diseased eye can be mistakenly diagnosed as normal. In this work, we propose a new image enhancement method to improve the quality of degraded images. A new method is used to enhance degraded-quality fundus images. In this method, the image is converted from the input RGB colour space to LAB colour space and then each normalized component is enhanced using contrast-limited adaptive histogram equalization. Human visual system (HVS)-based fundus image quality assessment, combined with diagnosis by experts, is used to evaluate the enhancement. The study included 191 degraded-quality fundus photographs of 143 subjects with optic media opacity. Objective quality assessment of image enhancement (range: 0-1) indicated that our method improved colour retinal image quality from an average of 0.0773 (variance 0.0801) to an average of 0.3973 (variance 0.0756). Following enhancement, area under curves (AUC) were 0.996 for the glaucoma classifier, 0.989 for the diabetic retinopathy (DR) classifier, 0.975 for the age-related macular degeneration (AMD) classifier and 0.979 for the other retinal diseases classifier. The relatively simple method for enhancing degraded-quality fundus images achieves superior image enhancement, as demonstrated in a qualitative HVS-based image quality assessment. This retinal image enhancement may, therefore, be employed to assist ophthalmologists in more efficient screening of retinal diseases and the development of computer-aided diagnosis.

Telemedicine screening of retinal diseases with a handheld portable non-mydriatic fundus camera

BackgroundWe modified and reconstructed a high image quality portable non-mydriatic fundus camera and compared it with the tabletop fundus camera to evaluate the efficacy of the new camera in detecting retinal diseases.MethodsWe designed and built a novel portable handheld fundus camera with telemedicine system. The image quality of fundus cameras was compared to that of existing commercial tabletop cameras by taking photographs of 364 eyes from the 254 patients. In all 800 fundus images taken by two camera types, 400 images per camera, were graded with the four image clarity classifications.ResultsUsing the portable fundus camera, 63% (252/400) images were graded as excellent overall quality, 20.5% (82/400) were good, 11.75% (47/400) were fair, and 4.75% (19/400) were inadequate. Using the tabletop fundus camera, 70.75% (283/400) images were graded as excellent overall quality, 20.4% (51/400) were good, 13.25% (53/400) were fair, and 3.25% (13/400) were inadequate. Common retinal diseases were easily identified from fundus images obtained from the portable fundus camera.ConclusionThe new type of non-mydriatic portable fundus camera was qualified to have professional quality of fundus images. The revolutionary screening camera provides a foundational platform which can potentially improve the accessibility of retinal screening programmes.

Human Visual System-Based Fundus Image Quality Assessment of Portable Fundus Camera Photographs.

Telemedicine and the medical "big data" era in ophthalmology highlight the use of non-mydriatic ocular fundus photography, which has given rise to indispensable applications of portable fundus cameras. However, in the case of portable fundus photography, non-mydriatic image quality is more vulnerable to distortions, such as uneven illumination, color distortion, blur, and low contrast. Such distortions are called generic quality distortions. This paper proposes an algorithm capable of selecting images of fair generic quality that would be especially useful to assist inexperienced individuals in collecting meaningful and interpretable data with consistency. The algorithm is based on three characteristics of the human visual system--multi-channel sensation, just noticeable blur, and the contrast sensitivity function to detect illumination and color distortion, blur, and low contrast distortion, respectively. A total of 536 retinal images, 280 from proprietary databases and 256 from public databases, were graded independently by one senior and two junior ophthalmologists, such that three partial measures of quality and generic overall quality were classified into two categories. Binary classification was implemented by the support vector machine and the decision tree, and receiver operating characteristic (ROC) curves were obtained and plotted to analyze the performance of the proposed algorithm. The experimental results revealed that the generic overall quality classification achieved a sensitivity of 87.45% at a specificity of 91.66%, with an area under the ROC curve of 0.9452, indicating the value of applying the algorithm, which is based on the human vision system, to assess the image quality of non-mydriatic photography, especially for low-cost ophthalmological telemedicine applications.

The fundus slit lamp.

Fundus biomicroscopy with the slit lamp as it is practiced widely nowadays was not established until the 1980-es with the introduction of the Volk lenses +90 and +60D. Thereafter little progress has been made in retinal imaging with the slit lamp. It is the aim of this paper to fully exploit the potential of a video slit lamp for fundus documentation by using easily accessible additions.Suitable still images are easily retrieved from videorecordings of slit lamp examinations. The effects of changements in the slit lamp itself (slit beam and apertures) and its examination equipment (converging lenses from +40 to +90D) on quality and spectrum of fundus images are demonstrated. Imaging software is applied for reconstruction of larger fundus areas in a mosaic pattern (Hugin®) and to perform the flicker test in order to visualize changes in the same fundus area at different points of time (Power Point®).The three lenses +90/+60/+40D are a good choice for imaging the whole spectrum of retinal diseases. Displacement of the oblique slit light can be used to assess changes in the surface profile of the inner retina which occurs e.g. in macular holes or pigment epithelial detachment. The mosaic function in its easiest form (one strip macula adapted to one strip with the optic disc) provides an overview of the posterior pole comparable to a fundus camera’s image. A reconstruction of larger fundus areas is feasible for imaging in vitreoretinal surgery or occlusive vessel disease. The flicker test is a fine tool for monitoring progressive glaucoma by changes in the optic disc, and it is also a valuable diagnostic tool in macular disease.Nearly all retinal diseases can be imaged with the slit lamp - irrespective whether they affect the posterior pole, mainly the optic nerve or the macula, the whole retina or only its periphery. Even a basic fundus controlled perimetry is possible. Therefore fundus videography with the slit lamp is a worthwhile approach especially for the many ophthalmologists without access to the most recent diagnostic equipment or a professional photographer at hand.Electronic supplementary materialThe online version of this article (doi:10.1186/s40064-015-0838-5) contains supplementary material, which is available to authorized users.

The effect of incorrect scanning distance on boundary detection errors and macular thickness measurements by spectral domain optical coherence tomography: a cross sectional study.

BackgroundTo investigate the influence of scan distance on retinal boundary detection errors (RBDEs) and retinal thickness measurements by spectral domain optical coherence tomography (SD-OCT).Methods10 eyes of healthy subjects, 10 eyes with diabetic macular edema (DME) and 10 eyes with neovascular age-related macular degeneration (AMD) were examined with RTVue SD-OCT. The MM5 protocol was used in two consecutive sessions to scan the macula. For the first session, the device was set 3.5 cm from the eye in order to obtain detectable signal with low fundus image quality (suboptimal setting) while in the second session a distance of 2.5 cm was set with a good quality fundus image. The signal strength (SSI) value was recorded. The score for retinal boundary detection errors (RBDE) was calculated for ten scans of each examination. RBDE scores were recorded for the whole scan and also for the peripheral 1.0 mm region. RBDE scores, regional retinal thickness values and SSI values between the two sessions were compared. The correlation between SSI and the number of RBDEs was also examined.ResultsThe SSI was significantly lower with suboptimal settings compared to optimal settings (63.9±12.0 vs. 68.3±12.2, respectively, p = 0.001) and the number of RBDEs was significantly higher with suboptimal settings in the “all-eyes” group along with the group of healthy subjects and eyes with DME (9.1±6.5 vs. 6.8±6.3, p = 0.007; 4.4±2.6 vs. 2.5±1.6, p = 0.035 and 9.7±3.3 vs. 5.1±3.7, p = 0.008, respectively). For these groups, significant negative correlation was found between the SSI and the number of RBDEs. In the AMD group, the number of RBDEs was markedly higher compared to the other groups and there was no difference in RBDEs between optimal and suboptimal settings with the errors being independent of the SSI. There were significantly less peripheral RBDEs with optimal settings in the “all-eyes” group and the DME subgroup (2.7±2.6 vs. 4.2±2.8, p = 0.001 and 1.4±1.7 vs. 4.1±2.2, p = 0.007, respectively). Retinal thickness in the two settings was significantly different only in the outer-superior region in DME.ConclusionsOptimal distance settings improve SD-OCT SSI with a decrease in RBDEs while retinal thickness measurements are independent of scanning distance.

Nonmydriatic Fundus Photography for Teleophthalmology Diabetic Retinopathy Screening in Rural and Urban Clinics

To evaluate the relative diagnostic value of nonmydriatic fundus photography (nFP) among patients screened for diabetic retinopathy in remote rural medical clinics and an urban academic medical center for nonadherence to recommended annual dilated eye examination. A retrospective cross-sectional study was performed among diabetic patients seen in primary outpatient clinics between 2006 and 2011 who were screened for diabetic retinopathy with nFP for history of nonadherence to recommended annual dilated eye examination. A single nonstereoscopic, 45°, 10-megapixel digital image of the disc and macula of both eyes was obtained locally and transmitted electronically to a retinal specialist for remote review. The results from remote rural Native American Indian reservations were compared with those from an urban academic family practice clinic. The proportion of subjects diagnosed with diabetic retinopathy and the quality of fundus images were compared. Among 872 patients (1,744 eyes) screened from rural sites and 517 subjects (1,034 eyes) screened from an urban site, images were of good quality for evaluation in 82.4% and 85.7% of subjects, respectively. Diabetic retinopathy was noted in 12.6% of rural subjects and 29.6% of urban subjects (p<0.001). nFP can be a useful tool in both rural and urban settings to screen for diabetic retinopathy in patients who are nonadherent to the recommended dilated annual eye exam. In our study population, a surprisingly higher percentage of diabetic subjects screened from the urban clinic had retinopathy compared with subjects screened in rural clinics.

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.

Der EYECUBE als 3-D-Multimedia-Imaging der Makuladiagnostik

In the new generation of EYECUBE devices, the angiography image and the OCT are included in a 3D illustration as an integration. Other diagnostic procedures such as autofluorescence and ICG can also be correlated to the OCT. The aim was to precisely classify various two-dimensional findings in relation to each other. The new generation of OCT devices enables imaging with a low incidence of motion artefacts with very good fundus image quality - and with that, permits a largely automatic classification. The feature enabling the integration of the EYECUBE was further developed with new software, so that not only the topographic image (red-free, autofluorescence) can be correlated to the Cirrus OCT, but also all other findings gathered within the same time frame can be correlated to each other. These were brightened and projected onto the cube surface in a defined interval. The imaging procedures can be selected in a menu toolbar. Topographic volumetry OCT images can be overlayed. The practical application of the new method was tested on patients with macular disorders. By lightening up the results from various diagnostic procedures, it is possible of late to directly compare pathologies to each other and to the OCT results. In all patients (n = 45 eyes) with good single-image quality, the automated integration into the EYECUBE was possible (to a great extent). The application is not dependent on a certain type of device used in the procedures performed. The increasing level of precision in imaging procedures and the handling of large data volumes has led to the possibility of examining each macular diagnostics procedure from the comparative perspective: imaging (photo) with perfusion (FLA, ICG) and morphology (OCT). The exclusion of motion artefacts and the reliable scan position in the course of the imaging process increases the informative value of OCT.

An Image Based Auto-Focusing Algorithm forDigital Fundus Photography

In fundus photography, the task of fine focusing the image is demanding and lack of focus is quite often the cause of suboptimal photographs. The introduction of digital cameras has provided an opportunity to automate the task of focusing. We have developed a software algorithm capable of identifying best focus. The auto-focus (AF) method is based on an algorithm we developed to assess the sharpness of an image. The AF algorithm was tested in the prototype of a semi-automated nonmydriatic fundus camera designed to screen in the primary care environment for major eye diseases. A series of images was acquired in volunteers while focusing the camera on the fundus. The image with the best focus was determined by the AF algorithm and compared to the assessment of two masked readers. A set of fundus images was obtained in 26 eyes of 20 normal subjects and 42 eyes of 28 glaucoma patients. The 95% limits of agreement between the readers and the AF algorithm were -2.56 to 2.93 and -3.7 to 3.84 diopter and the bias was 0.09 and 0.71 diopter, for the two readers respectively. On average, the readers agreed with the AF algorithm on the best correction within less than 3/4 diopter. The intraobserver repeatability was 0.94 and 1.87 diopter, for the two readers respectively, indicating that the limit of agreement with the AF algorithm was determined predominantly by the repeatability of each reader. An auto-focus algorithm for digital fundus photography can identify the best focus reliably and objectively. It may improve the quality of fundus images by easing the task of the photographer.

Die Bedeutung der digitalen Integrationsmethode (DIM) von OCT und FLA für die klinische Anwendung

We describe the establishment of a new digital integration method (DIM) for the anatomically precise and objective correlation of OCT and FLA. The aim of the study was to evaluate the clinical impact of DIM and deduce possible new knowledge in unclear angiographic findings. DIM uses a newly developed software for the integration of OCT and FLA images of the same date. Corresponding vascular bifurcations and OCT scan position are marked in OCT and FLA images. Pathological areas in OCT or FLA can be marked and compared. A fundus enhancement system (FES) was developed for a better fundus image quality in OCT. Using a retinal tracking system (RTS) eye movements can be compensated. DIM was evaluated in 225 patients with macular diseases. Using FES, the DIM was successful in 95 % of the patients in contrast to 25 % without FES. Using the video system of RTS eye movement artefacts could be compensated in all patients. Cases with clear angiographic findings showed a high correspondence (66 %) with the OCT findings. A discrepancy between FLA and OCT findings was found in 33 %, especially in leakages of unknown origin. DIM was of high value in age-related macular degeneration (AMD), detecting a heterogeneous morphology of serous, solid RPE detachments, retinal cystoid formation and choroidal neovascularisation (CNV) and differentiating it in a precise topography. The lesion size of classic CNV with a hypofluorescent halo was in OCT always larger than in FLA. Through the correction of the eye movement artefacts by DIM and FES, a fundamental weak point in the previous OCT pictures is counterbalanced. Using DIM the OCT scan position for the follow-up can now be assessed in a safe and reliable manner which is mandatory for the evidence of OCT studies. The detection of precise morphology (serous RPE detachments) in angiographically unclear phenomena can lead to treatment consequences, e. g., avoiding contraindicated areas in photodynamic therapy. Using the new morphological information new knowledge is obtained on uncertain angiographic phenomena by this integration method (DIM).

Image quality characteristics of a novel colour scanning digital ophthalmoscope (SDO) compared with fundus photography

To establish a set of quality parameters for grading image quality and apply those to evaluate the fundus image quality obtained by a new scanning digital ophthalmoscope (SDO) compared with standard slide photography. On visual analogue scales a total of eight image characteristics were defined: overall quality, contrast, colour brilliance, focus (sharpness), resolution and details, noise, artefacts and validity of clinical assessment. Grading was repeated after 4 months to assess repeatability. Fundus images of 23 patients imaged digitally by SDO and by Zeiss 450FF fundus camera using Kodak film were graded side-by-side by three graders. Lens opacity was quantified with the Interzeag Lens Opacity Meter 701. For all of the eight scales of image quality, good repeatability within the graders (mean Kendall's W 0.69) was obtained after 4 months. Inter-grader agreement ranged between 0.31 and 0.66. Despite the SDO's limited nominal image resolution of 720 x 576 pixels, the Zeiss FF 450 camera performed better in only two of the subscales - noise (p = 0.001) and artefacts (p = 0.01). Lens opacities significantly influenced only the two subscales 'resolution' and 'details', which deteriorated with increasing media opacities for both imaging systems. Distinct scales to grade image characteristics of different origin were developed and validated. Overall SDO digital imaging was found to provide fundus pictures of a similarly high level of quality as expert photography on slides.