Retinal vascular tree morphology: a semi-automatic quantification

Assessment of Energy Requirement for the Retinal Arterial Network in Normal and Hypertensive Subjects

The retinal arterial network structure can be altered by systemic diseases such as hypertension and diabetes. In order to compare the energy requirement for maintaining retinal blood flow and vessel wall metabolism between normal and hypertensive subjects, 3D hypothetical models of a representative retinal arterial bifurcation were constructed based on topological features derived from retinal images. Computational analysis of blood flow was performed, which accounted for the non-Newtonian rheological properties of blood and peripheral vessel resistance. The results suggested that the rate of energy required to maintain the blood flow and wall metabolism is much lower for normal subjects than for hypertensives, with the latter requiring 49.2% more energy for an entire retinal arteriolar tree. Among the several morphological factors, length-to-diameter ratio was found to have the most significant influence on the overall energy requirement.

A function for quality evaluation of retinal vessel segmentations

Retinal blood vessel assessment plays an important role in the diagnosis of ophthalmic pathologies. The use of digital images for this purpose enables the application of a computerized approach and has fostered the development of multiple methods for automated vascular tree segmentation. Metrics based on contingency tables for binary classification have been widely used for evaluating the performance of these algorithms. Metrics from this family are based on the measurement of a success or failure rate in the detected pixels, obtained by means of pixel-to-pixel comparison between the automated segmentation and a manually-labeled reference image. Therefore, vessel pixels are not considered as a part of a vascular structure with specific features. This paper contributes a function for the evaluation of global quality in retinal vessel segmentations. This function is based on the characterization of vascular structures as connected segments with measurable area and length. Thus, its design is meant to be sensitive to anatomical vascularity features. Comparison of results between the proposed function and other general quality evaluation functions shows that this proposal renders a high matching degree with human quality perception. Therefore, it can be used to enhance quality evaluation in retinal vessel segmentations, supplementing the existing functions. On the other hand, from a general point of view, the applied concept of measuring descriptive properties may be used to design specialized functions aimed at segmentation quality evaluation in other complex structures.

Analysis and algorithmic generation of hepatic vascular systems

A proper geometric model of the vascular systems in the liver is crucial for modeling blood flow, the connection between the organ and the rest of the organism. In vivo imaging does not provide sufficient details, so an algorithmic concept for extending measured vascular tree data is needed such that geometrically realistic structures can be generated. We develop a quantification of similarity in terms of different geometric features. This involves topological Strahler ordering of the vascular trees, statistical testing, and averaging. Invariant features are identified in human clinical in vivo CT scans. Results of the existing "Constrained Constructive Optimization" algorithm are compared to real vascular tree data. To improve bifurcation angles in the algorithmic results, we implement a postprocessing step calibrated to the measured features. This framework is finally applied to generate realistic additional details in a patient-specific hepatic vascular tree data set.

Automated discovery and quantification of image-based complex phenotypes: a twin study of drusen phenotypes in age-related macular degeneration

PURPOSE

Determining the relationships between phenotype and genotype of many disorders can improve clinical diagnoses, identify disease mechanisms, and enhance therapy. Most genetic disorders result from interaction of many genes that obscure the discovery of such relationships. The hypothesis for this study was that image analysis has the potential to enable formalized discovery of new visible phenotypes. It was tested in twins affected with age-related macular degeneration (AMD).

METHODS

Fundus images from 43 monozygotic (MZ) and 32 dizygotic (DZ) twin pairs with AMD were examined. First, soft and hard drusen were segmented. Then newly defined phenotypes were identified by using drusen distribution statistics that significantly separate MZ from DZ twins. The ACE model was used to identify the contributions of additive genetic (A), common environmental (C), and nonshared environmental (E) effects on drusen distribution phenotypes.

RESULTS

Four drusen distribution characteristics significantly separated MZ from DZ twin pairs. One encoded the quantity, and the remaining three encoded the spatial distribution of drusen, achieving a zygosity prediction accuracy of 76%, 74%, 68%, and 68%. Three of the four phenotypes had a 55% to 77% genetic effect in an AE model, and the fourth phenotype showed a nonshared environmental effect (E model).

CONCLUSIONS

Computational discovery of genetically determined features can reveal quantifiable AMD phenotypes that are genetically determined without explicitly linking them to specific genes. In addition, it can identify phenotypes that appear to result predominantly from environmental exposure. The approach is rapid and unbiased, suitable for large datasets, and can be used to reveal unknown phenotype-genotype relationships.

Automated measurement of the arteriolar-to-venular width ratio in digital color fundus photographs

A decreased ratio of the width of retinal arteries to veins [arteriolar-to-venular diameter ratio (AVR)], is well established as predictive of cerebral atrophy, stroke and other cardiovascular events in adults. Tortuous and dilated arteries and veins, as well as decreased AVR are also markers for plus disease in retinopathy of prematurity. This work presents an automated method to estimate the AVR in retinal color images by detecting the location of the optic disc, determining an appropriate region of interest (ROI), classifying vessels as arteries or veins, estimating vessel widths, and calculating the AVR. After vessel segmentation and vessel width determination, the optic disc is located and the system eliminates all vessels outside the AVR measurement ROI. A skeletonization operation is applied to the remaining vessels after which vessel crossings and bifurcation points are removed, leaving a set of vessel segments consisting of only vessel centerline pixels. Features are extracted from each centerline pixel in order to assign these a soft label indicating the likelihood that the pixel is part of a vein. As all centerline pixels in a connected vessel segment should be the same type, the median soft label is assigned to each centerline pixel in the segment. Next, artery vein pairs are matched using an iterative algorithm, and the widths of the vessels are used to calculate the AVR. We trained and tested the algorithm on a set of 65 high resolution digital color fundus photographs using a reference standard that indicates for each major vessel in the image whether it is an artery or vein. We compared the AVR values produced by our system with those determined by a semi-automated reference system. We obtained a mean unsigned error of 0.06 (SD 0.04) in 40 images with a mean AVR of 0.67. A second observer using the semi-automated system obtained the same mean unsigned error of 0.06 (SD 0.05) on the set of images with a mean AVR of 0.66. The testing data and reference standard used in this study has been made publicly available.

Temporary morphological changes in plus disease induced during contact digital imaging

OBJECTIVE

To compare and quantify the retinal vascular changes induced by non-intentional pressure contact by digital handheld camera during retinopathy of prematurity (ROP) imaging by means of a computer-based image analysis system, Retinal Image multiScale Analysis.

METHODS

A set of 10 wide-angle retinal pairs of photographs per patient, who underwent routine ROP examinations, was measured. Vascular trees were matched between 'compression artifact' (absence of the vascular column at the optic nerve) and 'not compression artifact' conditions. Parameters were analyzed using a two-level linear model for each individual parameter for arterioles and venules separately: integrated curvature (IC), diameter (d), and tortuosity index (TI).

RESULTS

Images affected with compression artifact showed significant vascular d (P<0.01) changes in both arteries and veins, as well as in artery IC (P<0.05). Vascular TI remained unchanged in both groups.

CONCLUSIONS

Non-adverted corneal pressure with the RetCam lens could compress and decrease intra-arterial diameter or even collapse retinal vessels. Careful attention to technique is essential to avoid absence of the arterial blood column at the optic nerve head that is indicative of increased pressure during imaging.

Plus Disease: Why is it Important in Retinopathy of Prematurity?

Retinopathy of prematurity (ROP) is one of the leading causes of preventable blindness in childhood. Early posterior pole vascular signs of severe ROP have been studied since the first description of the disease. The progressive changes that take place in the posterior pole vessels of an extremely premature baby occur in a predictable fashion soon after birth. These vascular changes are described as plus disease and are defined as abnormal dilation and tortousity of the blood vessels during ROP that may go on to total retinal detachment. The ophthalmological community now has a better understanding of the pathology and cascade of events taking place in the posterior pole of an eye with active ROP. Despite many years of scientific work on plus disease, there continue to be many challenges in defining the severity and quantification of the vascular changes. It is believed that understanding of the vascular phenomenons in patients with ROP will help in designing new treatment strategies that will help in salvaging many of the eyes with severe ROP.

Computerized analysis of narrow-field ROP images for the assessment of vessel caliber and tortuosity

Retinopathy of prematurity (ROP) is a disease involving abnormal development of retinal vasculature in premature infants, which might eventually lead to retinal detachment and visual loss. The quantitative assessment of vessel morphological features, such as width and tortuosity, can improve the clinical diagnosis and evaluation of ROP. We propose here a computerized system for the vascular analysis of narrow-field premature infant images. It is based on the manual drafting of the vessel axis, followed by automatic Canny filter edge extraction and automatic caliber and tortuosity estimation. We implemented this method as a web-based tool, ROPnet, which allows the quantitative assessment of vessel width and tortuosity simply using a web browser. To test the accuracy of the estimated parameters, fifteen narrow-field (30°) retinal images were acquired in infants with a non-contact fundus camera and analyzed with ROPnet. We compared the results with the corresponding ground-truth values derived from manual analysis. Average widths and tortuosities estimated with ROPnet vs. manual ones showed a correlation coefficient of 0.96 and 0.90, respectively.

A new supervised method for blood vessel segmentation in retinal images by using gray-level and moment invariants-based features

This paper presents a new supervised method for blood vessel detection in digital retinal images. This method uses a neural network (NN) scheme for pixel classification and computes a 7-D vector composed of gray-level and moment invariants-based features for pixel representation. The method was evaluated on the publicly available DRIVE and STARE databases, widely used for this purpose, since they contain retinal images where the vascular structure has been precisely marked by experts. Method performance on both sets of test images is better than other existing solutions in literature. The method proves especially accurate for vessel detection in STARE images. Its application to this database (even when the NN was trained on the DRIVE database) outperforms all analyzed segmentation approaches. Its effectiveness and robustness with different image conditions, together with its simplicity and fast implementation, make this blood vessel segmentation proposal suitable for retinal image computer analyses such as automated screening for early diabetic retinopathy detection.

Laser preconditioning on cranial bone site: analysis of morphological vascular parameters

BACKGROUND AND OBJECTIVES

Bone vascularization is a key factor in the bone healing process following X-ray irradiation. Preserving the vascular network from X-ray-induced injury is a relevant approach in the promotion of bone healing. Previously, we developed a protocol of laser preconditioning (810 nm diode laser, 36 J/cm²) prior to X-ray radiation (18.75 Gy) which protects the bone vascular network from deleterious effects of X-ray radiation. The aim of this present work is to characterize the effects of laser preconditioning on the bone through a morphological analysis of vascular parameters.

MATERIALS AND METHODS

Digital images of the vascular plexus were taken through an optical bone chamber which was implanted onto the calvaria of rabbits. Bespoke software was used for the quantification of the vessels (classified in four groups according to their diameter), vessel length, and number of nodes at weeks 0, 4, and 8. Twenty rabbits were divided into four groups: control group #1 (n = 5); laser group #2 (n = 5). X-ray radiation group #3 (n = 5), laser preconditioning 24 hours prior to X-ray radiation group #4 (n = 5).

RESULTS

The bone vascular network was stable for groups #1 and #2. Statistical analysis showed a significant reduction of each observed vascular parameter for groups #3 and #4. In the laser preconditioned group #4 the loss was less marked than in the X-ray group #3, especially for large vessels (diameter >50 µm).

DISCUSSION AND CONCLUSION

We provide in vivo microcirculatory evidence to support the concept of laser preconditioning of bone. A computer-based semi-automatic system is described to quantify superficial bone vascular network parameters that had been treated by laser preconditioning prior to X-ray radiation. Laser preconditioning significantly attenuates the deletion of the superficial bone vascular network irradiated by X-ray, especially concerning large diameter vessels.

Plus disease in retinopathy of prematurity: quantitative analysis of vascular change

PURPOSE

To examine the relationship between rate of vascular change and plus disease diagnosis.

DESIGN

Retrospective observational case-control study.

METHODS

Wide-angle images were taken bilaterally from 37 infants at 31 to 33 weeks and 35 to 37 weeks postmenstrual age (PMA). The semi-automated Retinal Image multiScale Analysis system was used to measure parameters for all arteries and veins: integrated curvature, diameter, and tortuosity index. A reference standard diagnosis (plus vs not plus) was defined for each eye by consensus of 5 experts at 35 to 37 weeks PMA. Weekly rate of change in parameters was compared in eyes with plus vs not plus disease. Receiver operating characteristic area under the curve (AUC) was calculated for plus disease detection based on 1) weekly rates of parameter change between 31 to 33 weeks and 35 to 37 weeks PMA and 2) parameter values at 35 to 37 weeks only.

RESULTS

Weekly rates of change in all venous parameters were significantly different in eyes with plus vs not plus disease, particularly for tortuosity index (P < .0004) and diameter (P = .018). Using weekly rate of change, AUC for plus disease detection was highest for venous tortuosity index (0.819) and venous diameter (0.712). Using the 35 to 37-week PMA image only, AUC was highest for venous integrated curvature (0.952) and diameter (0.789).

CONCLUSION

Rate of change in venous, but not arterial, parameters is correlated with plus disease development in this data set. This did not appear to contribute information beyond analysis of an image at 35 to 37 weeks PMA only.

2D Fast Vessel Visualization Using a Vessel Wall Mask Guiding Fine Vessel Detection

The paper addresses the fine retinal-vessel's detection issue that is faced in diagnostic applications and aims at assisting in better recognizing fine vessel anomalies in 2D. Our innovation relies in separating key visual features vessels exhibit in order to make the diagnosis of eventual retinopathologies easier to detect. This allows focusing on vessel segments which present fine changes detectable at different sampling scales. We advocate that these changes can be addressed as subsequent stages of the same vessel detection procedure. We first carry out an initial estimate of the basic vessel-wall's network, define the main wall-body, and then try to approach the ridges and branches of the vasculature's using fine detection. Fine vessel screening looks into local structural inconsistencies in vessels properties, into noise, or into not expected intensity variations observed inside pre-known vessel-body areas. The vessels are first modelled sufficiently but not precisely by their walls with a tubular model-structure that is the result of an initial segmentation. This provides a chart of likely Vessel Wall Pixels (VWPs) yielding a form of a likelihood vessel map mainly based on gradient filter's intensity and spatial arrangement parameters (e.g., linear consistency). Specific vessel parameters (centerline, width, location, fall-away rate, main orientation) are post-computed by convolving the image with a set of pre-tuned spatial filters called Matched Filters (MFs). These are easily computed as Gaussian-like 2D forms that use a limited range sub-optimal parameters adjusted to the dominant vessel characteristics obtained by Spatial Grey Level Difference statistics limiting the range of search into vessel widths of 16, 32, and 64 pixels. Sparse pixels are effectively eliminated by applying a limited range Hough Transform (HT) or region growing. Major benefits are limiting the range of parameters, reducing the search-space for post-convolution to only masked regions, representing almost 2% of the 2D volume, good speed versus accuracy/time trade-off. Results show the potentials of our approach in terms of time for detection ROC analysis and accuracy of vessel pixel (VP) detection.

Visual cycle modulation in neurovascular retinopathy

Rats with oxygen-induced retinopathy (OIR) model the pediatric retinal disease retinopathy of prematurity (ROP). Recent findings in OIR rats imply a causal role for the rods in the ROP disease process, although only experimental manipulation of rod function can establish this role conclusively. Accordingly, a visual cycle modulator (VCM) - with no known direct effect on retinal vasculature - was administered to "50/10 model" OIR Sprague-Dawley rats to test the hypotheses that it would 1) alter rod function and 2) consequently alter vascular outcome. Four litters of pups (N=46) were studied. For two weeks, beginning on postnatal day (P) 7, the first and fourth litters were administered 6 mg kg(-1) N-retinylacetamide (the VCM) intraperitoneally; the second and third litters received vehicle (DMSO) alone. Following a longitudinal design, retinal function was assessed by electroretinography (ERG) and the status of the retinal vessels was monitored using computerized fundus photograph analysis. Rod photoreceptor and post-receptor response amplitudes were significantly higher in VCM-treated than in vehicle-treated rats; deactivation of phototransduction was also significantly more rapid. Notably, the arterioles of VCM-treated rats showed significantly greater recovery from OIR. Presuming that the VCM did not directly affect the retinal vessels, a causal role for the neural retina - particularly the rod photoreceptors - in OIR was confirmed. There was no evidence of negative alteration of photoreceptor function consequent to VCM treatment. This finding implicates the rods as a possible therapeutic target in neurovascular diseases such as ROP.

Computational analysis of oxygen transport in the retinal arterial network

PURPOSE

The retina has a high oxygen consumption, making it particularly vulnerable to vascular insults, impairing oxygen and nutrient supply. The aim of this study was to develop a detailed computational model for quantitative analysis of blood flow and oxygen transport in physiologically realistic retinal arterial networks. Such a model will allow us to examine the effect of topological changes in retinal vasculature on hemodynamics and oxygen distribution in the retinal circulation.

MATERIALS AND METHODS

The Navier-Stokes equations for blood flow and the convection-diffusion equation for oxygen transfer were solved numerically to obtain detailed blood flow and oxygen distribution patterns in a retinal arterial tree. The geometrical outlines of the central retinal artery and its major branches were extracted from retinal images acquired from a healthy young adult by a Zeiss FF450+ fundus camera. The reconstructed subject-specific retinal arterial network geometry was combined with a structured tree model for the distal peripheral vessels. The non-Newtonian rheological properties of blood were incorporated by using an empirical viscosity model to account for the Fahraeus-Lindqvist effect.

RESULTS

The model predicted pressure drops in the range of 11-14.6 mmHg between the inlet and outlets of the reconstructed network and non-uniform oxygen tension, which varied with the vessel diameter and distance from the optic disc. The mean oxygen saturation in retinal arteries was 93.1% for vessels larger than 50 mum in diameter and 82.2% for smaller arterioles.

CONCLUSIONS

Our numerical results are in good agreement with in vivo measurements reported in the literature, demonstrating the potential of our model for prediction of oxygen distribution and intravascular oxygen tension profiles in the retinal arterial network. This paves the way for investigating the effects of parameter variation, simulating cases not available from experimental studies.

Vascular intersection detection in retina fundus images using a new hybrid approach

The use of vascular intersection aberration as one of the signs when monitoring and diagnosing diabetic retinopathy from retina fundus images (FIs) has been widely reported in the literature. In this paper, a new hybrid approach called the combined cross-point number (CCN) method able to detect the vascular bifurcation and intersection points in FIs is proposed. The CCN method makes use of two vascular intersection detection techniques, namely the modified cross-point number (MCN) method and the simple cross-point number (SCN) method. Our proposed approach was tested on images obtained from two different and publicly available fundus image databases. The results show a very high precision, accuracy, sensitivity and low false rate in detecting both bifurcation and crossover points compared with both the MCN and the SCN methods.

The neurovascular retina in retinopathy of prematurity

The continuing worldwide epidemic of retinopathy of prematurity (ROP), a leading cause of childhood visual impairment, strongly motivates further research into mechanisms of the disease. Although the hallmark of ROP is abnormal retinal vasculature, a growing body of evidence supports a critical role for the neural retina in the ROP disease process. The age of onset of ROP coincides with the rapid developmental increase in rod photoreceptor outer segment length and rhodopsin content of the retina with escalation of energy demands. Using a combination of non-invasive electroretinographic (ERG), psychophysical, and image analysis procedures, the neural retina and its vasculature have been studied in prematurely born human subjects, both with and without ROP, and in rats that model the key vascular and neural parameters found in human ROP subjects. These data are compared to comprehensive numeric summaries of the neural and vascular features in normally developing human and rat retina. In rats, biochemical, anatomical, and molecular biological investigations are paired with the non-invasive assessments. ROP, even if mild, primarily and persistently alters the structure and function of photoreceptors. Post-receptor neurons and retinal vasculature, which are intimately related, are also affected by ROP; conspicuous neurovascular abnormalities disappear, but subtle structural anomalies and functional deficits may persist years after clinical ROP resolves. The data from human subjects and rat models identify photoreceptor and post-receptor targets for interventions that promise improved outcomes for children at risk for ROP.

Comparison of the retinal microvasculature in European and African-Caribbean people with diabetes

Diabetes aggravates the impact of elevated BP (blood pressure) on the microcirculation, and people of African ancestry with diabetes are more susceptible to microvascular damage than Europeans. In the present study, we investigated possible differences in the retinal microcirculation in people of European and African-Caribbean ethnicity with diabetes that might account for this. A total of 51 subjects with Type 2 diabetes (age 40–65 years; 25 male; 29 African-Caribbean) were studied. Clinic and 24 h ambulatory BP, and fasting glucose, insulin and lipids were measured. Digital retinal images were analysed using custom-written semi-automatic software to determine: LDR (length/diameter ratio) and AVR (arteriolar/venular diameter ratio), branching angles, vessel tortuosity and NT (number of terminal vessel branches). Arterioles were narrower in European people with diabetes than in African-Caribbean people with diabetes [mean (S.D.) arteriolar diameter, 76 (7) compared with 82 (11) μm respectively (P=0.03); arteriolar LDR, 28.1 (8.5) compared with 23.7 (7.0) respectively (P=0.046); and AVR, 0.66 (0.21) compared with 0.90 (0.36) respectively (P=0.028)]. Ethnic differences in arteriolar LDR, arteriolar diameter and AVR were not explained by differences in BP, but were attenuated by adjustment for the duration of diabetes. There was no significant relationship between BP and arteriolar narrowing in the group as a whole, although the relationship between arteriolar LDR and systolic BP was stronger in Europeans than African-Caribbeans [β=0.08 (0.07) compared with β=0.03 (0.06); P=0.03]. In conclusion, in the presence of diabetes, a relationship between BP and retinal arteriolar diameter was not evident and implies impaired small artery remodelling in the presence of diabetes. African-Caribbean people with diabetes have wider retinal arterioles and this could contribute to enhanced microvascular damage in this ethnic group.

Digital image analysis of plus disease in retinopathy of prematurity

An accurate assessment of retinopathy of prematurity (ROP) is essential in ensuring correct and timely treatment of this potentially blinding condition. Current modes of assessment are based upon clinical grading by expert examination of retinal changes. However, this may be subjective, unreliable and difficult and there has been significant interest in alternative means of measurement. These have been made possible through technological advancements in image capture and analysis as well as progress in clinical research, highlighting the specific importance of plus disease in ROP. Progress in these two fields has highlighted the potential for digital image analysis of plus disease to be used as an objective, reliable and valid measurement of ROP. The potential for clinical and scientific advancement through this method is argued and demonstrated in this article. Along with the potential benefits, there are significant challenges such as in image capture, segmentation, measurement of vessel width and tortuosity; these are also addressed. After discussing and explaining the challenges involved, the research articles addressing digital image analysis of ROP are critically reviewed. Benefits and limitations of the currently published techniques for digital ROP assessment are discussed with particular reference to the validity and reliability of outcome measures. Finally, the general limitations of current methods of analysis are discussed and more diverse potential areas of development are discussed.

Unsupervised fuzzy based vessel segmentation in pathological digital fundus images

Performing the segmentation of vasculature in the retinal images having pathology is a challenging problem. This paper presents a novel approach for automated segmentation of the vasculature in retinal images. The approach uses the intensity information from red and green channels of the same retinal image to correct non-uniform illumination in color fundus images. Matched filtering is utilized to enhance the contrast of blood vessels against the background. The enhanced blood vessels are then segmented by employing spatially weighted fuzzy c-means clustering based thresholding which can well maintain the spatial structure of the vascular tree segments. The proposed method's performance is evaluated on publicly available DRIVE and STARE databases of manually labeled images. On the DRIVE and STARE databases, it achieves an area under the receiver operating characteristic curve of 0.9518 and 0.9602 respectively, being superior to those presented by state-of-the-art unsupervised approaches and comparable to those obtained with the supervised methods.

Systematic review of digital imaging screening strategies for retinopathy of prematurity

BACKGROUND

Retinal imaging with remote interpretation could decrease the number of diagnostic eye examinations that premature infants need for the detection of retinopathy of prematurity and thus decrease the time demand on the relatively small pool of ophthalmologists who perform retinopathy of prematurity examinations.

OBJECTIVE

Our goal was to review systematically the evidence regarding the reliability, validity, safety, costs, and benefits of retinal imaging to screen infants who are at risk for retinopathy of prematurity.

METHODS

We searched Medline, the Cochrane library, CINAHL, and the bibliographies of all relevant articles. All English-language studies regardless of design with primary data about our study questions were included. We excluded (1) studies that only included subjects with retinopathy of prematurity, (2) hypothetical models other than cost-effectiveness studies, and (3) validity studies without sufficient data to determine prevalence, sensitivity, and specificity or that only evaluated subjects for 1 component of retinopathy of prematurity (eg, plus disease only).

RESULTS

Studies of only 1 retinal imaging device (RetCam [Clarity Medical Systems, Inc, Pleasanton, CA]) met the inclusion criteria. There was a wide range in reported sensitivity, but specificity was high. There were several important limitations noted, including the eye as the unit of analysis instead of the individual or variations in the criteria for determining a true-positive or true-negative screening result. The risk of retinal hemorrhage resulting from imaging is low, and systemic effects (eg, bradycardia, hypertension, decreased oxygen saturation) are mild. No generalizable cost-effectiveness data were found.

CONCLUSIONS

The evidence base is not sufficient to recommend that retinal imaging be routinely adopted by NICUs to identify infants who have serious retinopathy of prematurity.

Retinal degenerative and hypoxic ischemic disease

A broad spectrum of retinal diseases affects both the retinal vasculature and the neural retina, including photoreceptor and postreceptor layers. The accepted clinical hallmarks of acute retinopathy of prematurity (ROP) are dilation and tortuosity of the retinal vasculature. Additionally, significant early and persistent effects on photoreceptor and postreceptor neural structures and function are demonstrated in ROP. In this paper, we focus on the results of longitudinal studies of electroretinographic (ERG) and vascular features in rats with induced retinopathies that model the gamut of human ROP, mild to severe. Two potential targets for pharmaceutical interventions emerge from the observations. The first target is immature photoreceptors because the status of the photoreceptors at an early age predicts later vascular outcome; this approach is appealing as it holds promise to prevent ROP. The second target is the interplay of the neural and vascular retinal networks, which develop cooperatively. Beneficial pharmaceutical interventions may be measured in improved visual outcome as well as lessening of the vascular abnormalities.

Plus disease in retinopathy of prematurity: development of composite images by quantification of expert opinion

PURPOSE

To demonstrate a methodology for generating composite wide-angle images of plus disease in retinopathy of prematurity (ROP), using quantitative analysis of expert opinions.

METHODS

Thirty-four wide-angle retinal images were independently interpreted by 22 ROP experts as "plus" or "not plus." All images were processed by the computer-based Retinal Image multiScale Analysis (RISA) system to calculate two parameters: arterial integrated curvature (AIC) and venous diameter (VD). Using a reference standard defined by expert consensus, sensitivity and specificity curves were calculated by varying the diagnostic cutoffs for AIC and VD. From these curves, individual vessels from multiple images were identified with particular diagnostic cutoffs, and were combined into composite wide-angle images using graphics-editing software.

RESULTS

The values associated with 75% underdiagnosis of true plus disease (i.e., 25% sensitivity cutoff) were AIC 0.061 and VD 4.272, the values associated with 50% underdiagnosis of true plus disease (i.e., a 50% sensitivity cutoff) were AIC 0.049 and VD 4.088, and the values associated with 25% underdiagnosis of true plus disease (i.e., 75% sensitivity cutoff) were AIC 0.042 and VD 3.795. Composite wide-angle images were generated by identifying and combining individual vessels with these characteristics.

CONCLUSIONS

Computer-based image analysis permits quantification of retinal vascular features, and a spectrum of abnormalities is seen in ROP. Selection of appropriate vessels from multiple images can produce composite plus disease images corresponding to expert opinions. This method may be useful for educational purposes, and for development of future disease definitions based on objective, quantitative principles.

A novel method for the automatic grading of retinal vessel tortuosity

Tortuosity is among the first alterations in the retinal vessel network to appear in many retinopathies, such as those due to hypertension. An automatic evaluation of retinal vessel tortuosity would help the early detection of such retinopathies. Quite a few techniques for tortuosity measurement and classification have been proposed, but they do not always match the clinical concept of tortuosity. This justifies the need for a new definition, able to express in mathematical terms the tortuosity as perceived by ophthalmologists. We propose here a new algorithm for the evaluation of tortuosity in vessels recognized in digital fundus images. It is based on partitioning each vessel in segments of constant-sign curvature and then combining together each evaluation of such segments and their number. The algorithm has been compared with other available tortuosity measures on a set of 30 arteries and one of 30 veins from 60 different images. These vessels had been preliminarily ordered by a retina specialist by increasing perceived tortuosity. The proposed algorithm proved to be the best one in matching the clinically perceived vessel tortuosity.

Retinal degeneration in children: dark adapted visual threshold and arteriolar diameter

To assess the condition of the retina in children with retinal degeneration due to Bardet-Biedl syndrome (BBS, n=41), Leber congenital amaurosis (LCA, n=31), or Usher syndrome (USH, n=13), the dark adapted visual threshold (DAT) and arteriolar diameters were measured. Compared to controls, the initial DATs of nearly all (83/85) were significantly elevated, and in 26/62 with serial DATs, significant progressive elevation occurred. Arteriolar diameters were significantly attenuated and narrowed with age in BBS and USH, but not LCA. Higher DATs were associated with narrower arterioles. Such non-invasive procedures can document the natural history of these retinal diseases and have the potential to assess response to future treatment.

Plus disease in retinopathy of prematurity: pilot study of computer-based and expert diagnosis

PURPOSE

To measure accuracy of plus disease diagnosis by recognized experts in retinopathy of prematurity (ROP), and to conduct a pilot study examining performance of a computer-based image analysis system, Retinal Image multiScale Analysis (RISA).

METHODS

Twenty-two ROP experts independently interpreted a set of 34 wide-angle retinal images for presence of plus disease. A reference standard diagnosis based on expert consensus was defined for each image. Images were analyzed by the computer-based system using individual and linear combinations of system parameters for arterioles and venules: integrated curvature (IC), diameter, and tortuosity index (TI). Sensitivity, specificity, and receiver operating characteristic areas under the curve (AUC) for plus disease diagnosis compared with the reference standard were determined for each expert, as well as for the computer-based system.

RESULTS

Expert sensitivity ranged from 0.308 to 1.000, specificity ranged from 0.571 to 1.000, and AUC ranged from 0.784 to 1.000. Among individual computer system parameters, venular IC had highest AUC (0.853). Among all computer system parameters, the linear combination of arteriolar IC, arteriolar TI, venular IC, venular diameter, and venular TI had highest AUC (0.967), which was greater than that of 18 (81.8%) of 22 experts.

CONCLUSIONS

Accuracy of ROP experts for plus disease diagnosis is imperfect. A computer-based image analysis system has potential to diagnose plus disease with high accuracy. Further research involving RISA system parameter cut-off values from this study are required to fully validate performance of this computer-based system compared with that of human experts.

Retinal blood vessel segmentation using line operators and support vector classification

In the framework of computer-aided diagnosis of eye diseases, retinal vessel segmentation based on line operators is proposed. A line detector, previously used in mammography, is applied to the green channel of the retinal image. It is based on the evaluation of the average grey level along lines of fixed length passing through the target pixel at different orientations. Two segmentation methods are considered. The first uses the basic line detector whose response is thresholded to obtain unsupervised pixel classification. As a further development, we employ two orthogonal line detectors along with the grey level of the target pixel to construct a feature vector for supervised classification using a support vector machine. The effectiveness of both methods is demonstrated through receiver operating characteristic analysis on two publicly available databases of color fundus images.

The clinical assessment of retinal microvascular structure and therapeutic implications

Examination of the retinal microvasculature is widely used to assess diabetic eye disease and as an indicator of target organ damage in hypertension. The diagnostic value of grading of hypertensive retinopathy is dubious; however, many recent studies have demonstrated that hypertensive retinopathy is associated with a range of risk factors for cardiovascular disease and may predict cardiovascular events independently of blood pressure. Developments in digital imaging and computer-assisted analysis have facilitated the quantitative assessment of microvascular changes in cardiovascular disease. These approaches may be useful for assessing cardiovascular risk and targeting therapeutic intervention.

Automated segmentation of retinal blood vessels and identification of proliferative diabetic retinopathy

Proliferative diabetic retinopathy can lead to blindness. However, early recognition allows appropriate, timely intervention. Fluorescein-labeled retinal blood vessels of 27 digital images were automatically segmented using the Gabor wavelet transform and classified using traditional features such as area, perimeter, and an additional five morphological features based on the derivatives-of-Gaussian wavelet-derived data. Discriminant analysis indicated that traditional features do not detect early proliferative retinopathy. The best single feature for discrimination was the wavelet curvature with an area under the curve (AUC) of 0.76. Linear discriminant analysis with a selection of six features achieved an AUC of 0.90 (0.73-0.97, 95% confidence interval). The wavelet method was able to segment retinal blood vessels and classify the images according to the presence or absence of proliferative retinopathy.

A new tracking system for the robust extraction of retinal vessel structure

Identification and measurement of blood vessels in retinal images could allow quantitative evaluation of clinical features, which may allow early diagnosis and effective monitoring of therapies in retinopathy. A new system is proposed for the automatic extraction of the vascular structure in retinal images, based on a sparse tracking technique. After processing pixels on a grid of rows and columns to determine a set of starting points (seeds), the tracking procedure starts. It moves along the vessel by analyzing subsequent vessel cross sections (lines perpendicular to the vessel direction), and extracting the vessel center, calibre and direction. Vessel points in a cross section are found by means of a fuzzy c-means classifier. When tracking stops because of a critical area, e.g. low contrast, bifurcation or crossing, a "bubble technique" module is run. It grows and analyzes circular scan lines around the critical points, allowing the exploration of the vessel structure beyond the critical areas. After tracking the vessels, identified segments are connected by a greedy connection algorithm. Finally bifurcations and crossings are identified analyzing vessel end points with respect to the vessel structure. Numerical evaluation of the performances of the system compared to human expert are reported.

Segmentation of blood vessels from red-free and fluorescein retinal images

The morphology of the retinal blood vessels can be an important indicator for diseases like diabetes, hypertension and retinopathy of prematurity (ROP). Thus, the measurement of changes in morphology of arterioles and venules can be of diagnostic value. Here we present a method to automatically segment retinal blood vessels based upon multiscale feature extraction. This method overcomes the problem of variations in contrast inherent in these images by using the first and second spatial derivatives of the intensity image that gives information about vessel topology. This approach also enables the detection of blood vessels of different widths, lengths and orientations. The local maxima over scales of the magnitude of the gradient and the maximum principal curvature of the Hessian tensor are used in a multiple pass region growing procedure. The growth progressively segments the blood vessels using feature information together with spatial information. The algorithm is tested on red-free and fluorescein retinal images, taken from two local and two public databases. Comparison with first public database yields values of 75.05% true positive rate (TPR) and 4.38% false positive rate (FPR). Second database values are of 72.46% TPR and 3.45% FPR. Our results on both public databases were comparable in performance with other authors. However, we conclude that these values are not sensitive enough so as to evaluate the performance of vessel geometry detection. Therefore we propose a new approach that uses measurements of vessel diameters and branching angles as a validation criterion to compare our segmented images with those hand segmented from public databases. Comparisons made between both hand segmented images from public databases showed a large inter-subject variability on geometric values. A last evaluation was made comparing vessel geometric values obtained from our segmented images between red-free and fluorescein paired images with the latter as the "ground truth". Our results demonstrated that borders found by our method are less biased and follow more consistently the border of the vessel and therefore they yield more confident geometric values.

Quantification of topological changes in retinal vascular architecture in essential and malignant hypertension

OBJECTIVE

Grading of hypertension severity by fundoscopic appearance is difficult and inaccurate. We investigated whether essential hypertension (EHT) and malignant phase hypertension (MHT) were associated with quantifiable abnormalities of the topology and architecture of the retinal circulation.

METHODS

The topology and architecture of the retinal microvasculature were compared in images from 20 normotensive subjects, 20 patients with EHT and 20 patients with MHT. Digitized retinal photographs were analysed by a novel multiscale image analysis method using a semi-automated program to quantify geometrical and topological properties of arteriolar and venular trees.

RESULTS

EHT was associated with an increase in the arteriolar length-to-diameter ratio (P < 0.01). There were also alterations in arteriolar topology indicative of rarefaction, including a marked reduction in the number of terminal branches in EHT (P < 0.01). These changes in the arteriolar network were exaggerated in MHT and there was also increased venular tortuosity and venular rarefaction in MHT compared with normotensive subjects.

CONCLUSIONS

Hypertension is associated with marked topological alterations in the retinal vasculature, and quantification of these changes may be a useful novel approach to the assessment of target organ damage in hypertension.

Segmentation of retinal blood vessels by combining the detection of centerlines and morphological reconstruction

This paper presents an automated method for the segmentation of the vascular network in retinal images. The algorithm starts with the extraction of vessel centerlines, which are used as guidelines for the subsequent vessel filling phase. For this purpose, 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. Our approach was tested on two publicly available databases and its results are compared with recently published methods. The results demonstrate that our algorithm outperforms other solutions and approximates the average accuracy of a human observer without a significant degradation of sensitivity and specificity.

Diagnosis of plus disease in retinopathy of prematurity using Retinal Image multiScale Analysis

PURPOSE

To evaluate a semiautomated image analysis software package, Retinal Image multiScale Analysis (RISA), for the diagnosis of plus disease in preterm infants with retinopathy of prematurity (ROP).

METHODS

Digital images of the posterior pole showing both disc and macula in preterm infants with ROP were analyzed with an enhanced version of RISA. Venules (N = 106) and arterioles (N = 44) were identified, and integrated curvature, diameter, and tortuosity of the vessels were calculated. After the RISA calculations were completed, the origins of the vessels were determined to be 32 eyes in 16 infants (12 eyes with plus disease, 20 with no plus disease, as diagnosed by ophthalmic examination). Vessels were sorted into two groups-plus disease and no plus disease-and each RISA parameter was compared using the Mann-Whitney test. For each parameter, sensitivity and specificity were plotted as a function of cutoff criterion, receiver operating characteristic (ROC) curves were constructed, and the areas under the curve (AUC) were calculated.

RESULTS

For both arterioles and venules, each of the three parameters was significantly larger for the plus disease group. For instance, the median estimated arteriolar and venular diameters were approximately 12 mum greater in plus disease. Sensitivity and specificity plots indicated good accuracy of each parameter for the diagnosis of plus disease. The AUC showed that curvature had the highest diagnostic accuracy (0.911 for arterioles, 0.824 for venules).

CONCLUSIONS

The strong performance of RISA parameters in this sample suggests that RISA may be useful for diagnosing plus disease in preterm infants with ROP.

Three-dimensional analysis of complex branching vessels in confocal microscopy images

The characteristic of confocal microscopy (CM) vascular data is that it contains many tiny vessels with branching and complex structure. In this work, an automated method for quantitative analysis and reconstruction of cerebral vessels from CM images is presented in which the extracted centerline of the vessels plays the key role. To assess the efficiency and accuracy of different centerline extraction methods, a comparison among three fully automated approaches is given. The centerline extraction methods studied in this work are a snake model, a path planning approach, and a distance transform-based method. To evaluate the accuracy of the quantitative parameters of vessels such as length and diameter, we apply the method to synthetic data. These results indicate that the snake model and the path planning method are more accurate in extracting the quantitative parameters. The efficiency of the approach in clinical applications is then confirmed by applying the method to real CM images. All three methods investigated in this work are accurate enough to correctly distinguish between normal and stroke brain data, while the snake model is the fastest for clinical applications. In addition, three-dimensional visualization, reconstruction, and characterization of CM vascular images of rat brains are presented.

3-D quantification and visualization of vascular structures from confocal microscopic images using skeletonization and voxel-coding

This paper presents an image processing approach for information extraction from three-dimensional (3-D) images of vasculature. It extracts quantitative information such as skeleton, length, diameter, and vessel-to-tissue ratio for different vessels as well as their branches. Furthermore, it generates 3-D visualization of vessels based on desired anatomical characteristics such as vessel diameter or 3-D connectivity. Steps of the proposed approach are: (1) pre-processing, (2) distance mappings, (3) branch labeling, (4) quantification, and (5) visualization. We have tested and evaluated the proposed algorithms using simulated images of multi-branch vessels and real confocal microscopic images of the vessels in rat brains. Experimental results illustrate performance of the methods and usefulness of the results for medical image analysis applications.