Observer sensitivity to retinal vessel diameter and tortuosity in retinopathy of prematurity: a model system

Retinopathy of Prematurity-assist: Novel Software for Detecting Plus Disease

Purpose

To design software with a novel algorithm, which analyzes the tortuosity and vascular dilatation in fundal images of retinopathy of prematurity (ROP) patients with an acceptable accuracy for detecting plus disease.

Methods

Eighty-seven well-focused fundal images taken with RetCam were classified to three groups of plus, non-plus, and pre-plus by agreement between three ROP experts. Automated algorithms in this study were designed based on two methods: the curvature measure and distance transform for assessment of tortuosity and vascular dilatation, respectively as two major parameters of plus disease detection.

Results

Thirty-eight plus, 12 pre-plus, and 37 non-plus images, which were classified by three experts, were tested by an automated algorithm and software evaluated the correct grouping of images in comparison to expert voting with three different classifiers, k-nearest neighbor, support vector machine and multilayer perceptron network. The plus, pre-plus, and non-plus images were analyzed with 72.3%, 83.7%, and 84.4% accuracy, respectively.

Conclusions

The new automated algorithm used in this pilot scheme for diagnosis and screening of patients with plus ROP has acceptable accuracy. With more improvements, it may become particularly useful, especially in centers without a skilled person in the ROP field.

The rate of change in retinal vessel width and tortuosity in eyes at risk for retinopathy of prematurity

PURPOSE

To describe the rate of change in retinal vessel width and tortuosity in eyes that develop treatment-requiring, or type 1, retinopathy of prematurity (ROP) versus eyes that do not develop type 1 ROP.

METHODS

Posterior poles of eyes of 41 infants at risk for ROP were imaged longitudinally with a 30° fundus camera. Retinal vessel width and tortuosity were measured with computer-assisted image analysis. The rate of change per day in width and tortuosity up to the development of most severe ROP was calculated from linear regression and eyes with (n = 10) and without type 1 ROP (n = 31) were compared.

RESULTS

Eyes that developed type 1 ROP had a greater rate of change in width for venules and 3 widest vessels (P < 0.0001), and a greater rate of change in tortuosity for arterioles and 3 most tortuous vessels (P < 0.0001) than eyes that did not develop type 1 ROP. These vessel parameters discriminate the 2 groups well (area under the ROC curve, 0.79-0.90). A combination of venular width and arteriolar tortuosity had the best discriminative ability (area under the ROC curve, 0.96).

CONCLUSIONS

In this pilot study, eyes that eventually developed type 1 ROP demonstrated a faster increase in width and tortuosity of retinal vessels compared with those that did not. Further study of the kinetics of retinal vascular change in a larger sample may allow for the earlier identification of vision-threatening ROP.

Digital image analysis in retinopathy of prematurity: a comparison of vessel selection methods

PURPOSE

To evaluate vessel selection methods to distinguish between eyes with and without retinopathy of prematurity (ROP) and between different stages of ROP when quantifying the associated vessel changes in width and tortuosity semiautomatically from digital retinal images.

METHODS

Color digital images from 75 infants screened for ROP were cropped to a standardized diameter of 240 pixels and evaluated by semiautomated vessel analysis software, Computer-Aided Image Analysis of the Retina (CAIAR), to measure retinal vessel width and tortuosity. Two methods of vessel selection were used: (1) clinical observer selecting the most prominent arteriole or venule in each retinal quadrant (4-vessel analysis) and then separately the 4 most prominent arterioles and venules from each quadrant (8-vessel analysis); (2) CAIAR selecting, regardless of retinal quadrant, the 4 widest or most tortuous arterioles or venules. Selected vessels were measured by CAIAR for tortuosity and width.

RESULTS

When comparing ROP stages, whether observer or CAIAR selected and whether 4 or 8 vessels were analyzed, we found that arteriolar tortuosity was significantly greater with advancing ROP stage for stage 0 versus stage 2; stage 0 or 1 versus stage 3; stages 1+2 combined versus stage 3; and stage 0 versus 1+2+3 combined (P < 0.01). Venular tortuosity was significantly greater with advancing ROP stage for stage 0 versus stage 3 and stage 0 versus stages 1 and 2+3 combined (P < 0.01). Width parameters did not help us to distinguish between stages.

CONCLUSIONS

Distinguishing between arterioles and venules is not necessary to differentiate stage 0 ROP from stage 2 or 3 ROP when one is using CAIAR. Tortuosity shows more promise than width at providing a reliable vessel parameter for distinguishing between eyes without and with ROP.

Evaluation of vascular disease progression in retinopathy of prematurity using static and dynamic retinal images

PURPOSE

To measure accuracy and speed for detection of vascular progression in retinopathy of prematurity (ROP) from serial images. Two strategies are compared: static side-by-side presentation and dynamic flickering of superimposed image pairs.

DESIGN

Prospective comparative study.

METHODS

Fifteen de-identified, wide-angle retinal image pairs were taken from infants who eventually developed plus disease. Image pairs representing vascular disease progression were taken ≥1 week apart, and control images without progression were taken on the same day. Dynamic flickering pairs were created by digital image registration. Ten experts independently reviewed each image pair on a secure website using both strategies, and were asked to identify progression or state that images were identical. Accuracy and speed were measured, using examination date and ophthalmoscopic findings as a reference standard.

RESULTS

Using static images, experts were accurate in a mean (%) ± standard deviation (SD) of 11.4 of 15 (76%) ± 1.7 image pairs. Using dynamic flickering images, experts were accurate in a mean (%) ± SD of 11.3 of 15 (75%) ± 1.7 image pairs. There was no significant difference in accuracy between these strategies (P = .420). Diagnostic speed was faster using dynamic flickering (24.7 ± 8.3 seconds) vs static side-by-side images (40.3 ± 18.3 seconds) (P = .002). Experts reported higher confidence when interpreting dynamic flickering images (P = .001).

CONCLUSIONS

Retinal imaging provides objective documentation of vascular appearance, with potentially improved ability to recognize ROP progression compared to standard ophthalmoscopy. Speed of identifying vascular progression was faster by review of dynamic flickering image pairs than by static side-by-side images, although there was no difference in accuracy.

From laboratory to clinic: the development of web-based tools for the estimation of retinal diagnostic parameters

Over the years, tools for the analysis of retinal images have been developed by several research groups but their usage has been mainly confined within the developing institutions. One possibility to foster their adoption is to develop them as web-based tools. We present here three such systems we recently developed. They are specifically focused on the estimation of retinal vascular parameters, such as arteriolar narrowing (AVR parameter), vessel tortuosity, and vessel caliber narrowing and tortuosity in retinopathy of prematurity (ROP) images. These systems have been successfully evaluated as regards their reliability and will soon be publicly available to interested health care providers.

Pulmonary arterial hypertension: what the large pulmonary arteries tell us

The morphology of the large intrapulmonary arteries (PAs) in pulmonary hypertension (PH) has received limited attention. Dilation, pruning, abrupt tapering, and tortuosity of PAs occur, but whether different patients have distinct PA phenotypes is unknown. Pulmonary arteriograms from 41 pediatric patients with PH were blindly reviewed by four experts who assigned each angiogram one of three designations: straight (S), tortuous (T), or ambiguous (A). Hemodynamic variables and outcomes were compared to the phenotypes. Thirty patients were either T (19) or S (11); 11 were A. The phenotypes were not associated with age. Tortuous patients had higher PA pressure and resistance than the S group and less likely to react to inhaled nitric oxide than S patients (p < 0.05). Clinical outcomes were similar for the three groups. Thus, in PH patients two subtypes of PA morphology can often be discerned, a reflection of variability in PA tortuosity. These morphological subtypes have differing hemodynamic characteristics. The mechanism(s) underlying these differences is unknown, but neither hydrodynamic factors nor duration of PH are fully explanatory. Because PA morphology might reveal information regarding the biology of pathological remodeling, it might prove enlightening to assess the large PA phenotype in future studies of PH.

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.

Agreement among pediatric ophthalmologists in diagnosing plus and pre-plus disease in retinopathy of prematurity

PURPOSE

Plus disease has become the major criterion for laser treatment in infants with retinopathy of prematurity (ROP), but its assessment is subjective. Our purpose was to compare quadrant-level and eye-level assessment of plus disease and pre-plus disease among 3 experienced ROP examiners and to report their rate of agreement.

METHODS

One hundred eighty-one high-quality RetCam images from premature infants were graded by 3 of the authors. Dilation and tortuosity were judged separately using a scale of normal or sufficiently abnormal to meet criteria for pre-plus or plus disease.

RESULTS

There was disagreement on the presence of plus disease for 18 images (10%), on tortuosity sufficient for plus disease (plus tortuosity) for 26 images (14%), and on dilation sufficient for plus disease (plus dilation) for 26 images (14%). Of 67 images judged to have pre-plus disease or worse, there was disagreement on the presence of plus disease for 18 images (27%), on plus tortuosity for 25 images (37%), and on plus dilation for 21 images (31%). For distinguishing plus or pre-plus disease from normal, there was disagreement on pre-plus tortuosity for 38 of 181 images (21%) and on pre-plus dilation for 58 of 181 images (32%).

CONCLUSIONS

Three experienced ROP examiners disagreed frequently on the diagnosis of plus or pre-plus disease when evaluating cropped clinical photographs of infants, many of which had borderline plus disease. Further study is required to determine the implications of these observations on clinical decision making.

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.

A pilot study using "ROPtool" to quantify plus disease in retinopathy of prematurity

PURPOSE

The accurate diagnosis of plus disease is critical to optimize the timing of laser treatment. Unfortunately, it is highly subjective and error-prone. "ROPtool" is a computer program that automatically traces retinal blood vessels and measures their tortuosity and dilation. Our aims were to pilot ROPtool, determine its reliability and validity, and establish appropriate numerical thresholds for plus and pre-plus disease.

METHODS

Twenty high-quality images of the posterior poles of premature infants were collected. Two of the authors (DKW and SFF) independently judged tortuosity and dilation separately as plus, pre-plus, or normal for each quadrant of each image. Disagreements were adjudicated, and the results were considered to be the standard for comparison to ROPtool. These two authors then separately used ROPtool to analyze the same 20 images.

RESULTS

For determination of tortuosity sufficient for plus disease, ROPtool interuser agreement was 95% (19/20), compared with 90% (18/20) agreement by investigator judgment. Eye-level (2 MDs x 20 eyes) sensitivity of ROPtool in detecting tortuosity sufficient for plus disease averaged 95% (21/22) and specificity averaged 78% (14/18). Quadrant-level (2 MDs x 20 eyes x 4 quadrants) sensitivity averaged 85% (66/78) and specificity averaged 77% (63/82). A numeric threshold for pre-plus disease equal to 70% of the average tortuosity of the standard photograph of plus disease resulted in mean sensitivity of 89% (103/116) and mean specificity of 82% (36/44) in distinguishing quadrant-level tortuosity sufficient for pre-plus disease or worse from normal.

CONCLUSIONS

ROPtool can reduce subjectivity and thereby enhance the evaluation of plus and pre-plus disease.

Retinal image analysis: concepts, applications and potential

As digital imaging and computing power increasingly develop, so too does the potential to use these technologies in ophthalmology. Image processing, analysis and computer vision techniques are increasing in prominence in all fields of medical science, and are especially pertinent to modern ophthalmology, as it is heavily dependent on visually oriented signs. The retinal microvasculature is unique in that it is the only part of the human circulation that can be directly visualised non-invasively in vivo, readily photographed and subject to digital image analysis. Exciting developments in image processing relevant to ophthalmology over the past 15 years includes the progress being made towards developing automated diagnostic systems for conditions, such as diabetic retinopathy, age-related macular degeneration and retinopathy of prematurity. These diagnostic systems offer the potential to be used in large-scale screening programs, with the potential for significant resource savings, as well as being free from observer bias and fatigue. In addition, quantitative measurements of retinal vascular topography using digital image analysis from retinal photography have been used as research tools to better understand the relationship between the retinal microvasculature and cardiovascular disease. Furthermore, advances in electronic media transmission increase the relevance of using image processing in 'teleophthalmology' as an aid in clinical decision-making, with particular relevance to large rural-based communities. In this review, we outline the principles upon which retinal digital image analysis is based. We discuss current techniques used to automatically detect landmark features of the fundus, such as the optic disc, fovea and blood vessels. We review the use of image analysis in the automated diagnosis of pathology (with particular reference to diabetic retinopathy). We also review its role in defining and performing quantitative measurements of vascular topography, how these entities are based on 'optimisation' principles and how they have helped to describe the relationship between systemic cardiovascular disease and retinal vascular changes. We also review the potential future use of fundal image analysis in telemedicine.

Retinal detachment associated with subthreshold retinopathy of prematurity

PURPOSE

To report a series of infants who progressed from mild retinopathy of prematurity (ROP) to severe ROP with retinal detachment without demonstrating detectable threshold disease.

METHODS

Between January 1993 and August 1998, seven infants at Oregon Health Sciences University, followed in accordance with the Cryotherapy for Retinopathy of Prematurity Study (CRYO-ROP) protocol, progressed to retinal detachment despite documentation that threshold had not been reached. This outlying subset of patients was analyzed and compared to the cohort in the CRYO-ROP study.

RESULTS

Six of 7 patients were male, 6 (86%) patients had symmetric disease, and all patients were born outside the study hospital. Mean birth-weight was 877 g and mean gestational age was 26 weeks. Mean postconceptual age at the time of retinal detachment was 41 weeks. Because of bilateral detachment in 3 patients, the total number of study eyes is 10. Failure to achieve threshold resulted from insufficient clock hours or insufficient stage in 2 eyes and lack of plus in 8 eyes. Zone I disease was present in 1 eye.

CONCLUSION

Rarely, despite adhering to ROP examination protocol, the retina may detach without demonstrating antecedent threshold disease. Very low birthweight is a factor that may lead to a less predictable course. This study found a lack of plus disease results in failure to reach threshold more often than the occurrence of insufficient clock hours of stage 3 disease. Further study is needed to determine if selected cases of subthreshold ROP may benefit from ablative therapy.

Prognostic significance of vascular dilation and tortuosity insufficient for plus disease in retinopathy of prematurity

BACKGROUND

Plus disease, one of the most important prognostic indicators in retinopathy of prematurity (ROP), is designated as present or absent. A grading system based on comparison with standard, high-quality color photographs may be useful to more accurately describe the spectrum of vascular dilation and tortuosity preceding plus disease, but it is of practical value only if it has prognostic significance. We hypothesized that grading of "pre-plus" vascular changes can identify eyes at risk for progression to vision-threatening ROP.

METHODS

Video clips of posterior pole images captured at the examination closest to 33 weeks' postconceptional age of 32 infants screened during an 18-month period were randomized. Two masked examiners viewed and graded the images in comparison with standard photographs representative of 5 distinct grades of retinal vascular dilation and tortuosity. A case-control design was used to compare the incidence of progression to stage 3 ROP, development of plus disease, and requirement of laser treatment between infants with normal posterior poles and those judged to have early dilation and tortuosity insufficient for plus disease.

RESULTS

Of the 8 patients with mild vascular dilation and tortuosity insufficient for plus disease, 5 (63%) eventually required laser treatment, 4 (50%) later developed stage 3 ROP, and 5 (63%) progressed to plus disease. Of the 24 patients with normal posterior poles, none required laser treatment, 2 (8%) developed stage 3 ROP, and none progressed to plus disease. The group with mild vascular dilation and tortuosity had a significantly higher incidence of progression to laser treatment (P =.0003), stage 3 ROP (P =.027), and plus disease (P =.0003).

CONCLUSIONS

Early vascular dilation and tortuosity judged insufficient for plus disease have prognostic significance in the early course of ROP. A grading system that uses standard, high-quality color photographs representing the spectrum of "pre-plus" vascular changes has potential utility in both the clinical and research settings.