Comparison of Spot Vision Screener and Tabletop Autorefractometer with Retinoscopy in the Pediatric Population

Objectives

Determining the accuracy of cycloplegic refractive error measurements made with the Spot Vision Screener (SVS, Welch Allyn Inc, Skaneateles Falls, NY, USA) is important for refractive assessment of uncooperative patients during optometric examinations. This study compared cycloplegic refractive errors measured by SVS and tabletop autorefractometer to cycloplegic retinoscopy in children.

Materials and Methods

Eighty-eight eyes of 44 subjects were examined in the study. Refractive error measurements were obtained under cycloplegia using retinoscopy, SVS, and Nidek ARK-530 tabletop autorefractometer (ARK-530, Nidek, Japan). Spherical and cylindrical values, spherical equivalents (SE), and Jackson cross-cylinder values at axes of 0° (J0) and 45° (J45) were recorded. Correlations between methods were analyzed using intraclass correlation coefficient (ICC) and Bland-Altman analysis.

Results

The mean age was 7 years (range: 6 months-17 years). Sixteen (36%) of the subjects were female and 28 (64%) were male. For SE there was excellent agreement between retinoscopy and SVS (ICC: 0.924) and between retinoscopy and tabletop autorefractometer (ICC: 0.995). While there was a moderate correlation between retinoscopy and SVS for cylindrical values (ICC: 0.686), excellent correlation was detected between retinoscopy and autorefractometer (ICC: 0.966). J0 and J45 crosscylinder power values were not correlated between retinoscopy and SVS (ICC: 0.472) or retinoscopy and tabletop autorefractometer (ICC: 0.442). Retinoscopy was correlated with both SVS and tabletop autorefractometer for all parameters within ±1.96 standard deviations in Bland-Altman analysis.

Conclusion

Cycloplegic retinoscopy is the gold standard for refractive error measurement in the pediatric population. However, it requires time and experienced professionals. This study revealed moderate to good agreement between SVS and retinoscopy, with better agreement in spherical errors than cylindrical errors. Although the SVS is intended for screening programs, it may also be useful in the pediatric eye office to estimate spherical refractive error in uncooperative patients.

Objective and subjective assessment of accommodative insufficiency

SIGNIFICANCE

A variety of subjective and objective procedures are available to measure the amplitude of accommodation. However, it is unclear whether the standard criterion of Hofstetter's minimum minus 2 D can be used to diagnose accommodative insufficiency with each of these techniques.

PURPOSE

The use of objective dynamic retinoscopy and three subjective techniques to diagnosis accommodative insufficiency was examined.

METHODS

A total of 632 subjects between 8 and 19 years of age were enrolled. Accommodative lag, monocular accommodative facility, and subjective (push-up, modified push-down, and minus lens) and objective (dynamic retinoscopy) amplitude of accommodation were quantified. Accommodative insufficiency was diagnosed based on Hofstetter's minimum minus 2 D for each subjective method, as well as adding an additional subjective criterion (either accommodative lag exceeding 0.75 D or monocular accommodative facility falling below the age-expected norms).

RESULTS

The prevalence of accommodative insufficiency was lowest and highest with the push-up (7.9 and 1%) and dynamic retinoscopy (94 and 12%) procedures when measured without and with the additional subjective criteria, respectively. Comparing the validity of dynamic retinoscopy against the traditional criterion, moderate to low sensitivity and high specificity were found. However, adding the additional subjective criteria improved the findings with moderate to high sensitivity and high specificity. Using a cutoff for dynamic retinoscopy of 7.50 D showed moderate diagnostic accuracy based on likelihood ratios.

CONCLUSIONS

It is clear that a revised definition of accommodative insufficiency is required, which must include the method of assessing accommodation. The various objective and subjective methods for quantifying the amplitude of accommodation are not interchangeable, and subjective assessment does not provide a valid measure of the accommodative response.

Comparison of the new self-contained darkroom refractive screener versus table-top autorefractor and cycloplegia retinoscopy in detecting refractive error

PURPOSE

By comparing the results of the new self-contained darkroom refractive screener (YD-SX-A) versus table-top autorefractor and cycloplegic retinoscopy, to evaluate the performance of the YD-SX-A in detecting refractive error in children and adolescents and then judge whether it can be used in refractive screening.

METHODS

Cross-sectional study. 1000 participants between the ages of 6 and 18 who visited the Optometry Center of the People's Hospital of Guangxi Zhuang Autonomous Region from June to December 2022 were selected. First, participants were instructed to measure their diopter with a table-top autorefractor (Topcon KR8800) and YD-SX-A in a noncycloplegic setting. After cycloplegia, they were retinoscopy by a professional optometrist. The results measured by three methods were collected respectively. To avoid deviation, only the right eye (1000 eyes) data were used in the statistical analysis. The Bland-Altman plots were used to evaluate the agreement of diopters measured by the three methods. The receiver operating characteristic (ROC) curves was used to analysis effectiveness of detecting refractive error of YD-SX-A.

RESULTS

The average age of participants was 10.77 ± 3.00 years, including 504 boys (50.4%) and 496 girls (49.6%). When YD-SX-A and cycloplegia retinoscopy (CR) were compared in the myopia group, there was no statistical difference in spherical equivalent (SE) (P > 0.05), but there was a statistical difference in diopter spherical (DS) and diopter cylinder (DC) (P < 0.05). Comparing the diopter results of Topcon KR8800 and CR, the difference between each test value in the myopia group was statistically significant (P < 0.05). In the hyperopia group, the comparison between YD-SX-A and CR showed no statistically significant differences in the DC (P > 0.05), but there were significant differences in the SE and DS (P < 0.05). In the astigmatism group, the SE, DS, and DC were statistically different, and the DC of YD-SX-A was lower than that of CR and Topcon KR8800. Bland-Altman plots indicated that YD-SX-A has a moderate agreement with CR and Topcon KR8800. The sensitivity and specificity of YD-SX-A for detecting myopia, hyperopia and astigmatism were 90.17% and 90.32%, 97.78% and 87.88%, 84.08% and 74.26%, respectively.

CONCLUSION

This study has identified that YD-SX-A has shown good performance in both agreement and effectiveness in detecting refractive error when compared with Topcon KR8800 and CR. YD-SX-A could be a useful tool for large-scale population refractive screening.

Training in retinoscopy: learning curves using a standardized method

BACKGROUND

Retinoscopy is one of the most effective objective techniques for evaluating refractive status, especially in non-cooperative patients. However, it presents a slow learning curve that often leads to student frustration. With the current Covid-19 pandemic and the need for social isolation, clinical education based on simulation has become more relevant. Therefore, we implemented retinoscopy laboratories and learning protocols to reduce student stress and learning time.

METHODS

We conducted a study to evaluate the retinoscopy learning curve using a new training protocol proposal. One hundred trainees were assessed in four stages, corresponding to 08, 12, 16, and 20 hours of training. Six different refractive defects were used trying to reproduce frequent conditions of care. The time spent on the assessment was not considered as additional training time. To analyze the data, we used non-parametric statistics and linear regression to assess the variables associated with training time and performance rate.

RESULTS

The mean performance score at 08 hrs was 32.49% (±16.69 SD); at 12 hrs was 59.75% (±18.80 SD); at 16 hrs was 70.83% (±18.53 SD) and at 20 hrs was 84.26% (±13.18 SD). Performance at 12 hrs was significative higher than 08 hrs of training, but did not show significant differences with the performance rate at 16 and 20 hrs. We found a strong positive correlation between performance and training time in retinoscopy (R = 0.9773, CI: 0.2678 - 0.9995 p = 0.0227).

CONCLUSION

This study showed that an increasing number of hours of practice positively correlates with performance in retinoscopy. The elaboration of a protocol and standardization of performance per hour also allowed us to estimate that a minimum of 13.4 hrs of practice is required to achieve 60% performance. Using the resulting formula, it is possible to determine the number of hours of retinoscopy practice are necessary to reach a certain level of performance.

Evaluation of a Pilot Protocol for Detecting Infant Hyperopia

SIGNIFICANCE

Highly hyperopic children are at greater risk for developing conditions such as strabismus, amblyopia, and early literacy and reading problems. High hyperopia is a common finding in infants in a pediatric medical practice, and early detection can be done effectively in that setting with tropicamide autorefraction.

PURPOSE

This study aimed to evaluate the effectiveness of a pilot screening program to detect high hyperopia in 2-month-old infants in a pediatric medical practice in Columbus, Ohio.

METHODS

Cycloplegic refractive error (1% tropicamide) was measured by retinoscopy and autorefraction with the Welch Allyn SureSight (Welch Allyn/Hillrom, Skaneateles Falls, NY) in 473 infants (55.4% female) who were undergoing their 2-month well-baby visit at their pediatrician's medical practice. Cycloplegic retinoscopy (1% cyclopentolate) was repeated at a subsequent visit in 35 infants with ≥+5.00 D hyperopia in the most hyperopic meridian during the screening.

RESULTS

Twenty-eight infants (5.9%) had high hyperopia (spherical equivalent, ≥+5.00 D), and 61 (12.9%) had high hyperopia (≥+5.00 D in at least one meridian of at least one eye) by retinoscopy with 1% tropicamide. The mean ± standard deviation spherical equivalent tropicamide cycloplegic refractive error measured with retinoscopy was +2.54 ± 1.54 D (range, -3.25 to +7.00 D) and with SureSight was +2.29 ± 1.64 D (range, -2.90 to +7.53 D). Retinoscopy done using 1% cyclopentolate was 0.44 ± 0.54 D more hyperopic in spherical equivalent than with 1% tropicamide ( P < .001).

CONCLUSIONS

High hyperopia was a common finding in 2-month-old infants in a pediatric medical setting that could be detected effectively by cycloplegic autorefraction using tropicamide. Greater cooperation between pediatric primary vision and medical care could lead to effective vision screenings designed to detect high hyperopia in infants.

Cycloplegic Autorefraction as a Substitute for Cycloplegic Retinoscopy in the Pediatric Population

PURPOSE

To evaluate whether cycloplegic autorefraction can provide similar results as cycloplegic retinoscopy, allowing more comprehensive ophthalmologists to be comfortable in managing pediatric refractive error and refractive amblyopia.

METHODS

This retrospective chart review was performed to determine the mean difference in sphere, cylinder, and axis between cycloplegic autorefraction and retinoscopy, both of which were obtained on the same eye at least 30 minutes after cycloplegia and dilation with a mixed solution of tropicamide, cyclopentolate, and phenylephrine.

RESULTS

A total of 34 eyes (18 right, 16 left) from 18 patients were included in the analysis. Mean sphere difference between cycloplegic autorefraction and retinoscopy was 0.044 ± 0.278 diopters (D) (95% CI: -1.275 to 1.363 D), mean cylinder difference was -0.081 ± 0.236 D (95% CI: -0.706 to 0.544 D), and mean axis difference was 7.059 ± 19.676 degrees (95% CI: -32.527 to 38.878 degrees). Mean differences in sphere, cylinder, and axis were not statistically significant (P = .362, .0541, and .377, respectively).

CONCLUSIONS

In this small sample population, cycloplegic autorefraction was comparable to cycloplegic retinoscopy. Recognition of amblyopia should still prompt evaluation by a pediatric ophthalmologist. Further research is necessary to confirm whether uncomplicated refractive error in children may be sufficiently detected and managed by a comprehensive ophthalmologist. [J Pediatr Ophthalmol Strabismus. 2022:59(6):422-427.].

Rise of the Machines? Comparison of Cycloplegic Refraction Using Retinoscopy and the Retinomax K-Plus 5 in Children

PURPOSE

To compare the ability to detect refractive anomalies in children using automated refraction versus retinoscopic cycloplegic refraction.

METHODS

A pediatric population from a pediatric eye institute underwent complete ophthalmic examinations. Children were randomly assigned to one of two pediatric optometrists who performed manual cycloplegic refraction using retinoscopy and automated cycloplegic refraction using a handheld autorefractometer (Retinomax K-plus 5; Right Mfg. Co., Ltd.). Recorded patient data included refraction values for each eye (sphere, astigmatism, and axis), use of glasses, and degree of cooperation.

RESULTS

Two hundred thirteen children were included. The mean age was 6.2 years. For all ages, strong associations were found in sphere and spherical equivalent (SE) measurements between the two methods (b = 0.78, P < .001; b = 0.71, P < .001; respectively). Among children older than 5 years, associations between the two methods were significant in all parameters (sphere: b = 0.99, P < .001; astigmatism: b = 0.69, P < .001; axis: b = 0.19, P < .05; SE: b = 0.97, P < .001), whereas among children 5 years and younger, a significant association was found only in the axis measurements (b = 0.31, P < .01). Retinomax K-plus 5 measurements showed significantly more hyperopic results in sphere measurements and higher astigmatism in all children examined, but this difference was markedly higher in children 5 years and younger. Good cooperation was observed in 94.1% of children older than 5 years and 77% of children 5 years and younger (P < .001).

CONCLUSIONS

The Retinomax K-plus 5 may be used for screening in children older than 5 years. However, in all age groups, it may not be accurate enough for treatment and decision making, even with good cooperation. [J Pediatr Ophthalmol Strabismus. 2022;59(6):380-387.].

Evaluation of a low-resource screening strategy for ophthalmic pathologies and associated neurological morbidity in an older Tanzanian HIV-positive population

Globally, 43 million people are living with HIV, 90% in developing countries. Increasing life expectancy with combination antiretroviral therapy (cART) results in chronic complications, including HIV-associated neurocognitive disorders (HAND) and eye diseases. HAND screening is currently challenging. Our aim was to evaluate clinical utility of retinopathy as a screening measure of HAND in older cART-treated individuals in Tanzania and feasibility of smartphone-based retinal screening in this low-resource setting. A cross-sectional systematic sample aged ≥ 50-years attending routine HIV follow-up in Tanzania were comprehensively assessed for HAND by American Academy of Neurology criteria and received ophthalmic assessment including smartphone-based retinal imaging. HAND and ophthalmic assessments were independent and blinded. Diagnostic accuracy was evaluated by AUROC curves. Of 129 individuals assessed, 69.8% were visually impaired. Thirteen had retinopathy. HAND prevalence was 66.7%. Retinopathy was significantly associated with HAND but HIV-disease factors (CD4, viral load) were not. Diagnostic accuracy of retinopathy for HAND was poor (AUROC 0.545-0.617) but specificity and positive predictive value were high. We conclude that ocular pathology and HAND appear highly prevalent in this low-resource setting. Although retinal screening cannot be used alone identify HAND, prioritization of individuals with abnormal retinal screening is a potential strategy in low-resource settings.

Risk factors and pregnancy outcomes associated with retinopathy in patients presenting with severe preeclampsia: A retrospective cohort study

The visual system was reported to be affected in over half of patients with preeclampsia (PE), though fundus examination was performed only among patients complaining of visual symptoms. Delayed diagnosis and treatment of PE-related retinopathy may lead to permanent visual impairment. Therefore, we hypothesize that some clinical or laboratory parameters could predict severity of retinal damage.The aim of the study was to explore the risk factors for retinopathy in severe preeclampsia (sPE) and investigate pregnancy outcomes with different degrees of retinopathy.This retrospective cohort study included women with sPE who underwent ophthalmoscopy and delivered after admission to West China Second University Hospital, between June 2013 and December 2016. Clinical and laboratory characteristics were retrieved from medical records. Patients confirmed with retinopathy were followed up with telephones. Multiple logistic regression analysis was performed to identify risk factors of PE-related retinopathy.Five hundred thirty-four patients were included, of which 17.6% having stage-1/2 retinopathy, 14.6% having stage-3/4 retinopathy, and 67.8% having normal retina. Compared with patients without retinopathy, patients with stage 3/4 retinopathy were more likely to have preterm-birth and low-birth-weight babies. Significant risk factors for stage 3/4 retinopathy in sPE included severe hypertension (odds ratio [OR] 2.24, 95% confidence interval [CI]: 1.10-4.56), elevated white blood cell (WBC) counts (OR 1.88, 95% CI: 1.05-3.35), decreased platelet counts (OR 2.12, 95% CI: 1.07-4.48), lactate dehydrogenase (LDH) concentration of >800 IU/L (OR 2.31, 95% CI: 1.05-5.06), low hemoglobin (HGB) concentrations of <110 g/L (OR 3.73, 95% CI: 1.21-11.47), 24-hour proteinuria of 2 to 5 g (OR 6.39, 95% CI: 2.84-14.39), and >5 g (OR 8.66, 95% CI: 3.67-20.44).This study confirms the association between retinopathy and preterm-birth and low-birth weight in sPE. The risk factors for severe PE-related retinopathy, including severe hypertension, platelet and WBC count, HGB and LDH concentration, and proteinuria, are associated with the development of retinopathy. Routine and repeated fundus examination is recommended for maternal monitoring in sPE.

Assessing amblyopia treatment using multifocal visual evoked potentials

BACKGROUND

To evaluate the effect of occlusion treatment for anisometropic amblyopia using multifocal visual evoked potentials (mfVEPs).

METHODS

The patients for this study comprised 19 patients (mean age 6.05 ± 1.65 years) with anisometropic amblyopia underwent mfVEP analysis using the RETIscan® system before and after occlusion treatment. After dividing the area into six ring areas and four quadrants, we analyzed the amplitudes and latencies of the mfVEPs.

RESULTS

The amplitudes of ring 1 (central field) in amblyopic eyes after treatment were significantly higher than those in the other rings (p = 0.001). The mfVEP amplitudes in each of the six rings between amblyopic eyes and fellow eyes at diagnosis and after occlusion treatment showed no significant differences. In quadrant 1 the amplitudes of the amblyopic eyes and fellow eyes were significantly different at the time of diagnosis (p = 0.005), whereas after occlusion treatment there was no significant difference (p = 0.888). The amplitudes for each of the six rings at diagnosis and after occlusion treatment in amblyopic eyes versus fellow eyes showed no significant difference. There were also no differences in the amplitudes in each of the four quadrants at the time of diagnosis and after occlusion treatment in amblyopic eyes versus fellow eyes. No significant difference was found in the comparison of latency values in each of the six rings or in each of the four quadrants at diagnosis and after occlusion treatment in amblyopic eyes versus their fellow eyes.

CONCLUSIONS

The amplitudes of quadrant 1 in amblyopic eyes compared with those of the fellow eyes at diagnosis were increased after occlusion treatment. Changes of the difference between amblyopic eyes and fellow eyes in quadrant 1 after occlusion treatment could be a useful, objective method for monitoring improvement in visual acuity.

The Role of Dynamic Retinoscopy in Predicting Infantile Accommodative Esotropia and Influencing Emmetropization

INTRODUCTION AND PURPOSE

To investigate the role that dynamic retinoscopy can play in reducing the occurrence of infantile accommodative esotropia and facilitating emmetropization in infants with high hyperopia.

PATIENTS AND METHODS

211 orthotropic infants under 1 year of age (3.5 to 12 months) identified as having ≥5 diopters of hyperopia in their more hyperopic eye on a routine eye exam. On enrollment, infants underwent an assessment of accommodation using dynamic retinoscopy as well as a cycloplegic refraction. Infants who showed normal accommodation were followed without spectacles. If dynamic retinoscopy showed subnormal accommodation, partial hyperopic correction that allowed for full binocular accommodative responses at near were prescribed. Main outcome measures were the occurrence of esotropia, changes in refractive error, and visual acuity.

RESULTS

Of the 211 infants enrolled, 146 showed normal accommodation and were followed without glasses (Group 1). None of these patients developed strabismus. Sixty-five patients showed subnormal accommodation and received partial hyperopic correction (Group 2). Thirty-four of the 65 (52%) in Group 2 did not develop strabismus (Group 2A) and 31 of the 65 (48%) developed strabismus (Group 2B). All 3 groups showed a reduction of hyperopia of 0.37D ± 0.25/year, 0.50D ± 0.28/year, and 0.60D ± 0.20/year in groups 1, 2A, and 2B, respectively. None of the differences between groups were statistically significant.

CONCLUSIONS

Normal accommodation on dynamic retinoscopy in orthotropic hyperopic infants is a predictor of continued good alignment and such infants can be followed without spectacles. Partial spectacle correction based on dynamic retinoscopy may have a beneficial effect on reducing the development of strabismus without impeding emmetropization. Early binocular accommodative behavior seems to be predictive of infants at risk of developing strabismus.

Analysis of the Ocular Refractive State in Fighting Bulls: Astigmatism Prevalence

The purpose of this study was to describe the ocular refractive state (ORS) of fighting bulls. The study consisted of 90 ophthalmological healthy animals (85 in post-mortem and 5 in living conditions, resp.). The ORS of the eyes (2 per animal) was determined using streak retinoscopy. In vivo animals were assessed at a fighting bull farm facility. Post-mortem measurements were carried out at a local arena. The ORS along the horizontal meridian ranged between -1.00 and +2.50 diopters (D), with a mean of +0.66 ± 0.85 D in post-mortem animals. Values for in vivo conditions were similar (+0.75 ± 0.46 D). Left and right eyes were highly correlated in both sets (p < 0.001). A fairly good correlation was also observed when comparing living and post-mortem eyes in the same animals. Anisometropia ≥ 1.00 D was diagnosed in 3 animals. Astigmatism (≥+0.5 D) was detected in 93% of the eyes. To our knowledge, the ORS of the fighting bull has been reported for the first time. Although values vary among individuals, all eyes presented a marked astigmatism. Whereas the horizontal meridian was slightly hyperopic, the vertical meridian was always closer to emmetropia. These results represent a starting point to understand the ocular optics of this kind of animals, which might benefit the selection of animals at the farm before being sent to the bullfighting arena.

[Optic disc swelling : A compilation of relevant differential diagnoses]

Optic disc oedema describes blurred optic disc margins. It describes an unspecific clinical sign with various possible underlying causal mechanisms. The correct diagnosis and the detection of the underlying disease is however of special clinical importance. This article summarizes the most important differential diagnoses and gives advice to find the correct diagnosis.

Image processing based automatic diagnosis of glaucoma using wavelet features of segmented optic disc from fundus image

Glaucoma is a disease of the retina which is one of the most common causes of permanent blindness worldwide. This paper presents an automatic image processing based method for glaucoma diagnosis from the digital fundus image. In this paper wavelet feature extraction has been followed by optimized genetic feature selection combined with several learning algorithms and various parameter settings. Unlike the existing research works where the features are considered from the complete fundus or a sub image of the fundus, this work is based on feature extraction from the segmented and blood vessel removed optic disc to improve the accuracy of identification. The experimental results presented in this paper indicate that the wavelet features of the segmented optic disc image are clinically more significant in comparison to features of the whole or sub fundus image in the detection of glaucoma from fundus image. Accuracy of glaucoma identification achieved in this work is 94.7% and a comparison with existing methods of glaucoma detection from fundus image indicates that the proposed approach has improved accuracy of classification.

An adaptive threshold based image processing technique for improved glaucoma detection and classification

Glaucoma is an optic neuropathy which is one of the main causes of permanent blindness worldwide. This paper presents an automatic image processing based method for detection of glaucoma from the digital fundus images. In this proposed work, the discriminatory parameters of glaucoma infection, such as cup to disc ratio (CDR), neuro retinal rim (NRR) area and blood vessels in different regions of the optic disc has been used as features and fed as inputs to learning algorithms for glaucoma diagnosis. These features which have discriminatory changes with the occurrence of glaucoma are strategically used for training the classifiers to improve the accuracy of identification. The segmentation of optic disc and cup based on adaptive threshold of the pixel intensities lying in the optic nerve head region. Unlike existing methods the proposed algorithm is based on an adaptive threshold that uses local features from the fundus image for segmentation of optic cup and optic disc making it invariant to the quality of the image and noise content which may find wider acceptability. The experimental results indicate that such features are more significant in comparison to the statistical or textural features as considered in existing works. The proposed work achieves an accuracy of 94.11% with a sensitivity of 100%. A comparison of the proposed work with the existing methods indicates that the proposed approach has improved accuracy of classification glaucoma from a digital fundus which may be considered clinically significant.

Automatic segmentation of the optic nerve head for deformation measurements in video rate optical coherence tomography

Optical coherence tomography (OCT) imaging has become a standard diagnostic tool in ophthalmology, providing essential information associated with various eye diseases. In order to investigate the dynamics of the ocular fundus, we present a simple and accurate automated algorithm to segment the inner limiting membrane in video-rate optic nerve head spectral domain (SD) OCT images. The method is based on morphological operations including a two-step contrast enhancement technique, proving to be very robust when dealing with low signal-to-noise ratio images and pathological eyes. An analysis algorithm was also developed to measure neuroretinal tissue deformation from the segmented retinal profiles. The performance of the algorithm is demonstrated, and deformation results are presented for healthy and glaucomatous eyes.

Investigating the influence of chromatic aberration and optical illumination bandwidth on fundus imaging in rats

Abstract. Rodent models are indispensable in studying various retinal diseases. Noninvasive, high-resolution retinal imaging of rodent models is highly desired for longitudinally investigating the pathogenesis and therapeutic strategies. However, due to severe aberrations, the retinal image quality in rodents can be much worse than that in humans. We numerically and experimentally investigated the influence of chromatic aberration and optical illumination bandwidth on retinal imaging. We confirmed that the rat retinal image quality decreased with increasing illumination bandwidth. We achieved the retinal image resolution of 10  μm using a 19 nm illumination bandwidth centered at 580 nm in a home-built fundus camera. Furthermore, we observed higher chromatic aberration in albino rat eyes than in pigmented rat eyes. This study provides a design guide for high-resolution fundus camera for rodents. Our method is also beneficial to dispersion compensation in multiwavelength retinal imaging applications.

Vessel extraction from non-fluorescein fundus images using orientation-aware detector

The automatic extraction of blood vessels in non-fluorescein eye fundus images is a tough task in applications such as diabetic retinopathy screening. However, vessel shapes have complex variations, and accurate modeling of retinal vascular structures is challenging. We have therefore developed a new approach to accurately extract blood vessels in non-fluorescein fundus images using an orientation-aware detector (OAD). The detector was designed according to the intrinsic property of vessels being locally oriented and having linearly elongated structures. We employ the OAD to extract vessel shapes with no assumptions on parametric orientations of vessel shapes. The orientations of vessels can be efficiently modeled by the energy distribution of Fourier transformation. Accordingly, both wide and thin vessels can be extracted with two-scale segmentation in which line operators are applied in large scale and the Gabor filter bank is applied in small scale. A post-processing technique, based on the path opening operation, is applied to eliminate false responses to nonvascular areas, such as retinal structures (optic disc and macula) and pathologies (exudates, hemorrhages,and microaneurysms). This makes the detector robust and structure-aware. By achieving a competitive CAL measurement of 80.82% for the DRIVE database and 68.94% for the STARE, the experimental results demonstrated that the OAD approach outperforms existing segmentation methods. Furthermore, the proposed approach effectively works with non-fluorescein fundus images and proves highly accurate and robust in complicated regions such as the central reflex, close vessels, and crossover points, despite a high level of illumination noise in the original data.

A multiscale decomposition approach to detect abnormal vasculature in the optic disc

This paper presents a multiscale method to detect neovascularization in the optic disc (NVD) using fundus images. Our method is applied to a manually selected region of interest (ROI) containing the optic disc. All the vessels in the ROI are segmented by adaptively combining contrast enhancement methods with a vessel segmentation technique. Textural features extracted using multiscale amplitude-modulation frequency-modulation, morphological granulometry, and fractal dimension are used. A linear SVM is used to perform the classification, which is tested by means of 10-fold cross-validation. The performance is evaluated using 300 images achieving an AUC of 0.93 with maximum accuracy of 88%.

Photoscreeners in the pediatric eye office: compared testability and refractions on high-risk children

PURPOSE

To compare refractive data and testability of Spot (PediaVision) and Plusoptix A09 (Plusoptix, Inc) photoscreeners and to compare each device with traditional cycloplegic retinoscopy.

DESIGN

Prospective, interventional case series.

METHODS

After informed consent, patients underwent testing with the Spot and Plusoptix photoscreeners before their examination by a pediatric ophthalmologist masked to the results. Data including testability and estimated refractions were entered into a Research Electronic Data Capture database for statistical analysis.

RESULTS

A total of 265 children were enrolled (mean age, 6.0 ± 3.4 years). Both devices produced a computer printout result in 250 (94.3%) of the patients. The Spot photoscreener provided a refractive estimate in all computer printouts, whereas the Plusoptix, used binocularly, provided a refractive estimate in 75.2% (188/250) of the printouts. Compared with cycloplegic retinoscopy, both devices underestimated hyperopia or overestimated myopia (-1.35 diopters [D] and -0.64 D, Spot and Plusoptix, respectively) and overestimated astigmatism (0.36 D and 0.32 D, Spot and Plusoptix, respectively). The intraclass correlation coefficient for spherical equivalents indicated good agreement between cycloplegic retinoscopy and Spot (0.806) and excellent agreement between cycloplegic retinoscopy and Plusoptix (0.898).

CONCLUSIONS

The Spot photoscreener provided refractive data on a greater percentage of children. The photorefractors correlated with cycloplegic retinoscopy refractive findings for sphere and spherical equivalents, but underestimated hyperopia or overestimated myopia and overestimated astigmatism. The binocular refractions of Plusoptix agreed more closely with the refractions of our pediatric ophthalmologists.

Exact surface registration of retinal surfaces from 3-D optical coherence tomography images

Nonrigid registration of optical coherence tomography (OCT) images is an important problem in studying eye diseases, evaluating the effect of pharmaceuticals in treating vision loss, and performing group-wise cross-sectional analysis. High dimensional nonrigid registration algorithms required for cross-sectional and longitudinal analysis are still being developed for accurate registration of OCT image volumes, with the speckle noise in images presenting a challenge for registration. Development of algorithms for segmentation of OCT images to generate surface models of retinal layers has advanced considerably and several algorithms are now available that can segment retinal OCT images into constituent retinal surfaces. Important morphometric measurements can be extracted if accurate surface registration algorithm for registering retinal surfaces onto corresponding template surfaces were available. In this paper, we present a novel method to perform multiple and simultaneous retinal surface registration, targeted to registering surfaces extracted from ocular volumetric OCT images. This enables a point-to-point correspondence (homology) between template and subject surfaces, allowing for a direct, vertex-wise comparison of morphometric measurements across subject groups. We demonstrate that this approach can be used to localize and analyze regional changes in choroidal and nerve fiber layer thickness among healthy and glaucomatous subjects, allowing for cross-sectional population wise analysis. We also demonstrate the method's ability to track longitudinal changes in optic nerve head morphometry, allowing for within-individual tracking of morphometric changes. This method can also, in the future, be used as a precursor to 3-D OCT image registration to better initialize nonrigid image registration algorithms closer to the desired solution.

Landmark detection for fusion of fundus and MRI toward a patient-specific multimodal eye model

Ophthalmologists typically acquire different image modalities to diagnose eye pathologies. They comprise, e.g., Fundus photography, optical coherence tomography, computed tomography, and magnetic resonance imaging (MRI). Yet, these images are often complementary and do express the same pathologies in a different way. Some pathologies are only visible in a particular modality. Thus, it is beneficial for the ophthalmologist to have these modalities fused into a single patient-specific model. The goal of this paper is a fusion of Fundus photography with segmented MRI volumes. This adds information to MRI that was not visible before like vessels and the macula. This paper contributions include automatic detection of the optic disc, the fovea, the optic axis, and an automatic segmentation of the vitreous humor of the eye.

Factors influencing prevalence of vision and ocular abnormalities among Jewish and Arab Israeli schoolchildren

BACKGROUND

The detection and correction of refractive errors is one of the priorities of the World Health Organization Initiative Vision 2020.

OBJECTIVES

To determine the factors related to ocular abnormalities (poor vision, refractive error or other abnormality) among schoolchildren in northern Israel.

METHODS

A cross-sectional population-based study was conducted among 2113 students aged 6-7 and 13-14 years old in 70 schools in northern Israel. Medical examination included vision history, clinical eye examination and vision testing. If a parent's informed consent was available, eye drops (cycloplegia) were delivered for fundus and retinoscopy testing. An ophthalmologist was asked to determine the need for the child's referral for further diagnostic procedures, treatment and/or follow-up. Multivariate analysis was limited to 1708 children, using data pertaining to the ophthalmologist's decision regarding referral, as well as vision and retinoscopy results.

RESULTS

Vision and/or ocular abnormality was prevalent in 21.5% (95% confidence interval 17.4-26.6%), predominantly among 13-14 year olds and Jewish children. Abnormal clinical findings were found in 5.7% of the students. Retinoscopy showed a higher prevalence of hypermetropia among 6-7 year olds and a higher prevalence of myopia and astigmatism among the 13-14 year olds. The multivariate analysis suggests an independent effect of retinoscopy abnormality (odds ratio = 3.85), vision abnormality (OR = 2.42), Jewish ethnicity (OR = 1.62) and 13-14 year old age group (OR = 1.26) on the ophthalmologist's referral decision.

CONCLUSIONS

Vision and/or ocular abnormality is an important health problem among schoolchildren in northern Israel. The independent effect of ethnicity and age on the ophthalmologist's referral decision should be further explored.

User-guided segmentation for volumetric retinal optical coherence tomography images

Despite the existence of automatic segmentation techniques, trained graders still rely on manual segmentation to provide retinal layers and features from clinical optical coherence tomography (OCT) images for accurate measurements. To bridge the gap between this time-consuming need of manual segmentation and currently available automatic segmentation techniques, this paper proposes a user-guided segmentation method to perform the segmentation of retinal layers and features in OCT images. With this method, by interactively navigating three-dimensional (3-D) OCT images, the user first manually defines user-defined (or sketched) lines at regions where the retinal layers appear very irregular for which the automatic segmentation method often fails to provide satisfactory results. The algorithm is then guided by these sketched lines to trace the entire 3-D retinal layer and anatomical features by the use of novel layer and edge detectors that are based on robust likelihood estimation. The layer and edge boundaries are finally obtained to achieve segmentation. Segmentation of retinal layers in mouse and human OCT images demonstrates the reliability and efficiency of the proposed user-guided segmentation method.

Landmark matching based retinal image alignment by enforcing sparsity in correspondence matrix

Retinal image alignment is fundamental to many applications in diagnosis of eye diseases. In this paper, we address the problem of landmark matching based retinal image alignment. We propose a novel landmark matching formulation by enforcing sparsity in the correspondence matrix and offer its solutions based on linear programming. The proposed formulation not only enables a joint estimation of the landmark correspondences and a predefined transformation model but also combines the benefits of the softassign strategy (Chui and Rangarajan, 2003) and the combinatorial optimization of linear programming. We also introduced a set of reinforced self-similarities descriptors which can better characterize local photometric and geometric properties of the retinal image. Theoretical analysis and experimental results with both fundus color images and angiogram images show the superior performances of our algorithms to several state-of-the-art techniques.

Magnetic resonance imaging of the retina: from mice to men

This mini-review provides an overview of magnetic resonance imaging (MRI) applications to study rodent, cat, non-human primate, and human retinas. These techniques include T(1) - and T(2) -weighted anatomical, diffusion, blood flow, blood volume, blood-oxygenation level dependent, manganese-enhanced, physiological, and functional MRI. Applications to study the retinas in diabetic retinopathy, glaucoma, and retinal degeneration are also reviewed. MRI offers some unique advantages compared with existing imaging techniques and has the potential to further our understanding of physiology and function in healthy and diseased retinas.

An automatic graph-based approach for artery/vein classification in retinal images

The classification of retinal vessels into artery/vein (A/V) is an important phase for automating the detection of vascular changes, and for the calculation of characteristic signs associated with several systemic diseases such as diabetes, hypertension, and other cardiovascular conditions. This paper presents an automatic approach for A/V classification based on the analysis of a graph extracted from the retinal vasculature. The proposed method classifies the entire vascular tree deciding on the type of each intersection point (graph nodes) and assigning one of two labels to each vessel segment (graph links). Final classification of a vessel segment as A/V is performed through the combination of the graph-based labeling results with a set of intensity features. The results of this proposed method are compared with manual labeling for three public databases. Accuracy values of 88.3%, 87.4%, and 89.8% are obtained for the images of the INSPIRE-AVR, DRIVE, and VICAVR databases, respectively. These results demonstrate that our method outperforms recent approaches for A/V classification.

Repeat Retinomax screening changes positive predictive value

PURPOSE

To quantify changes in autorefraction measurement with repeated readings using the Retinomax autorefractor and to investigate the clinical implications of the results.

METHODS

Children referred from a preschool vision screening program for a failed autorefraction screening test received repeat autorefraction as well as a comprehensive eye examination with cycloplegic retinoscopy at later follow-up. The intraclass correlation coefficient between initial and follow-up autorefraction was calculated to quantify changes in repeated measurements to determine whether the second autorefraction significantly changed the predictive value that a referred child would meet case definition. Cases were defined by AAPOS Vision Screening Committee amblyogenic risk factors under cycloplegic retinoscopy.

RESULTS

Repeat Retinomax autorefraction had an intraclass correlation of 0.70 in the right eye and 0.70 in the left eye for mean sphere. Of 636 children who were referred on their initial screening, 169 (26.5%) passed a repeat screening and this subpopulation had 7 cases (4.1%). Of the 467 (73.5%) who again met referral criteria at repeat screening, 268 (57.4%) met case definition. The difference in case rates between these subgroups was highly significant (Fisher exact test, P < 0.001).

CONCLUSIONS

There was clinically significant variability when autorefraction measurements were repeated among those referred from initial screening, allowing further risk stratification. In our study cohort, few children who passed repeat screening required further examination. Significant money and overtreatment risk may potentially be avoided by rescreening children who are initially referred from screening evaluations and only examining those who meet referral criteria after a second screening.

Rhegmatogenous retinal detachment--an ophthalmologic emergency

BACKGROUND

Rhegmatogenous retinal detachment is the most common retinological emergency threatening vision, with an incidence of 1 in 10 000 persons per year, corresponding to about 8000 new cases in Germany annually. Without treatment, blindness in the affected eye may result.

METHOD

Selective review of the literature.

RESULTS

Rhegmatogenous retinal detachment typically presents with the perception of light flashes, floaters, or a "dark curtain." In most cases, the retinal tear is a consequence of degeneration of the vitreous body. Epidemiologic studies have identified myopia and prior cataract surgery as the main risk factors. Persons in the sixth and seventh decades of life are most commonly affected. Rhegmatogenous retinal detachment is an emergency, and all patients should be seen by an ophthalmologist on the same day that symptoms arise. The treatment consists of scleral buckle, removal of the vitreous body (vitrectomy), or a combination of the two. Anatomical success rates are in the range of 85% to 90%. Vitrectomy is followed by lens opacification in more than 70% of cases. The earlier the patient is seen by an ophthalmologist, the greater the chance that the macula is still attached, so that visual acuity can be preserved.

CONCLUSION

Rhegmatogenous retinal detachment is among the main emergency indications in ophthalmology. In all such cases, an ophthalmologist must be consulted at once.

Computer aided diagnosis system for retinal analysis: automatic assessment of the vascular tortuosity

The tortuosity of a vessel, that is, how many times a vessel curves, and how these turns are, is an important value for the diagnosis of certain diseases. Clinicians analyze fundus images manually in order to estimate it, but there is many drawbacks as it is a tedious, time-consuming and subjective work. Thus, automatic image processing methods become a necessity, as they make possible the efficient computation of objective parameters. In this paper we will discuss Sirius (System for the Integration of Retinal Images Understanding Service), a web-based application that enables the storage and treatment of various types of diagnostic tests and, more specifically, its tortuosity calculation module.

Generalizing the majority voting scheme to spatially constrained voting

Generating ensembles from multiple individual classifiers is a popular approach to raise the accuracy of the decision. As a rule for decision making, majority voting is a usually applied model. In this paper, we generalize classical majority voting by incorporating probability terms pn,k to constrain the basic framework. These terms control whether a correct or false decision is made if k correct votes are present among the total number of n. This generalization is motivated by object detection problems, where the members of the ensemble are image processing algorithms giving their votes as pixels in the image domain. In this scenario, the terms pn,k can be specialized by a geometric constraint. Namely, the votes should fall inside a region matching the size and shape of the object to vote together. We give several theoretical results in this new model for both dependent and independent classifiers, whose individual accuracies may also differ. As a real world example, we present our ensemble-based system developed for the detection of the optic disc in retinal images. For this problem, experimental results are shown to demonstrate the characterization capability of this system. We also investigate how the generalized model can help us to improve an ensemble with extending it by adding a new algorithm.

Multiparametric optical coherence tomography imaging of the inner retinal hemodynamic response to visual stimulation

The hemodynamic response to neuronal activation is a well-studied phenomenon in the brain, due to the prevalence of functional magnetic resonance imaging. The retina represents an optically accessible platform for studying lamina-specific neurovascular coupling in the central nervous system; however, due to methodological limitations, this has been challenging to date. We demonstrate techniques for the imaging of visual stimulus-evoked hyperemia in the rat inner retina using Doppler optical coherence tomography (OCT) and OCT angiography. Volumetric imaging with three-dimensional motion correction, en face flow calculation, and normalization of dynamic signal to static signal are techniques that reduce spurious changes caused by motion. We anticipate that OCT imaging of retinal functional hyperemia may yield viable biomarkers in diseases, such as diabetic retinopathy, where the neurovascular unit may be impaired.

Automated characterization of blood vessels as arteries and veins in retinal images

In recent years researchers have found that alternations in arterial or venular tree of the retinal vasculature are associated with several public health problems such as diabetic retinopathy which is also the leading cause of blindness in the world. A prerequisite for automated assessment of subtle changes in arteries and veins, is to accurately separate those vessels from each other. This is a difficult task due to high similarity between arteries and veins in addition to variation of color and non-uniform illumination inter and intra retinal images. In this paper a novel structural and automated method is presented for artery/vein classification of blood vessels in retinal images. The proposed method consists of three main steps. In the first step, several image enhancement techniques are employed to improve the images. Then a specific feature extraction process is applied to separate major arteries from veins. Indeed, vessels are divided to smaller segments and feature extraction and vessel classification are applied to each small vessel segment instead of each vessel point. Finally, a post processing step is added to improve the results obtained from the previous step using structural characteristics of the retinal vascular network. In the last stage, vessel features at intersection and bifurcation points are processed for detection of arterial and venular sub trees. Ultimately vessel labels are revised by publishing the dominant label through each identified connected tree of arteries or veins. Evaluation of the proposed approach against two different datasets of retinal images including DRIVE database demonstrates the good performance and robustness of the method. The proposed method may be used for determination of arteriolar to venular diameter ratio in retinal images. Also the proposed method potentially allows for further investigation of labels of thinner arteries and veins which might be found by tracing them back to the major vessels.

Comparison between refractometer and retinoscopy in determining refractive errors in children--false doubt

Early detection of a refractive error and its correction are extremely important for the prevention of amblyopia (poor vision). The golden standard in the detection of refractive errors is retinoscopy--a method where the pupils are dilated in order to exclude accomodation. This results in a more accurate measurement of a refractive error. Automatic computer refractometer is also in use. The study included 30 patients, 15 boys, 15 girls aged 4-16. The first examination was conducted with refractometer on narrow pupils. Retinoscopy, followed by another examination with refractometer was performed on pupils dilated with mydriatic drops administered 3 times. The results obtained with three methods were compared. They indicate that in narrow pupils the autorefractometer revealed an increased diopter value in nearsightedness (myopia), the minus overcorrection, whereas findings obtained with retinoscopy and autorefractometer in mydriasis cycloplegia, were much more accurate. The results were statistically processed, which confirmed the differences between obtained measurements. These findings are consistent with the results of studies conducted by other authors. Automatic refractometry on narrow pupils has proven to be a method for detection of refractive errors in children. However, the exact value of the refractive error is obtained only in mydriasis--with retinoscopy or an automatic refractometer on dilated pupils.

Retinal microaneurysm detection through local rotating cross-section profile analysis

A method for the automatic detection of microaneurysms (MAs) in color retinal images is proposed in this paper. The recognition of MAs is an essential step in the diagnosis and grading of diabetic retinopathy. The proposed method realizes MA detection through the analysis of directional cross-section profiles centered on the local maximum pixels of the preprocessed image. Peak detection is applied on each profile, and a set of attributes regarding the size, height, and shape of the peak are calculated subsequently. The statistical measures of these attribute values as the orientation of the cross-section changes constitute the feature set that is used in a naïve Bayes classification to exclude spurious candidates. We give a formula for the final score of the remaining candidates, which can be thresholded further for a binary output. The proposed method has been tested in the Retinopathy Online Challenge, where it proved to be competitive with the state-of-the-art approaches. We also present the experimental results for a private image set using the same classifier setup.

Understanding the digital divide in the clinical setting: the technology knowledge gap experienced by US safety net patients during teleretinal screening

Differential access to everyday technology and healthcare amongst safety net patients is associated with low technological and health literacies, respectively. These low rates of literacy produce a complex patient "knowledge gap" that influences the effectiveness of telehealth technologies. To understand this "knowledge gap", six focus groups (2 African-American and 4 Latino) were conducted with patients who received teleretinal screenings in U.S. urban safety-net settings. Findings indicate that patients' "knowledge gap" is primarily produced at three points: (1) when patients' preexisting personal barriers to care became exacerbated in the clinical setting; (2) through encounters with technology during screening; and (3) in doctor-patient follow-up. This "knowledge gap" can produce confusion and fear, potentially affecting patients' confidence in quality of care and limiting their disease management ability. In rethinking the digital divide to include the consequences of this knowledge gap faced by patients in the clinical setting, we suggest that patient education focus on both their disease and specific telehealth technologies deployed in care delivery.

Visualization of three-dimensional ultra-high resolution OCT in virtual reality

Three-dimensional reconstruction of optical coherence tomography (OCT) images is a modern technique that helps interpret the images and understand the underlying disease. However, the 3D reconstruction displayed on commercial devices is of limited quality: images are shown on 2D screens and it is difficult or impossible to adjust the view point and capture the data set from a meaningful perspective. We did a preliminary study to evaluate the applicability of a novel, 3D TV-based virtual reality system with interactive volume rendering software to clinical diagnostics and present a workflow, which can incorporate virtual reality technology at various levels of immersion into the daily medical practice, from interactive VR systems to printed media.

Wavefront aberration reconstruction from tangential refractive powers measured with spatial dynamic skiascopy

The aim of this work was to study, using numerical simulations, the attainable level of accuracy to reconstruct the wavefront aberrations from tangential refractive power data measured with dynamic skiascopy. Two mathematical methods have been implemented. The first one is based on curve integration of the curvature data, previously interpolated with cubic splines. The second one reconstructs the three-dimensional wavefront surface, represented by a Zernike polynomial expansion, using a two-step least-squares method. The different factors affecting the reconstruction--noise, sampling, and wavefront patterns--were quantified. The results provide useful information to design more efficient experimental setups based on spatial dynamic skiascopy.

Simple methods for segmentation and measurement of diabetic retinopathy lesions in retinal fundus images

Diabetic retinopathy (DR) is one of the most important complications of diabetes mellitus, which causes serious damages in the retina, consequently visual loss and sometimes blindness if necessary medical treatment is not applied on time. One of the difficulties in this illness is that the patient with diabetes mellitus requires a continuous screening for early detection. So far, numerous methods have been proposed by researchers to automate the detection process of DR in retinal fundus images. In this paper, we developed an alternative simple approach to detect DR. This method was built on the inverse segmentation method, which we suggested before to detect Age Related Macular Degeneration (ARMDs). Background image approach along with inverse segmentation is employed to measure and follow up the degenerations in retinal fundus images. Direct segmentation techniques generate unsatisfactory results in some cases. This is because of the fact that the texture of unhealthy areas such as DR is not homogenous. The inverse method is proposed to exploit the homogeneity of healthy areas rather than dealing with varying structure of unhealthy areas for segmenting bright lesions (hard exudates and cotton wool spots). On the other hand, the background image, dividing the retinal image into high and low intensity areas, is exploited in segmentation of hard exudates and cotton wool spots, and microaneurysms (MAs) and hemorrhages (HEMs), separately. Therefore, a complete segmentation system is developed for segmenting DR, including hard exudates, cotton wool spots, MAs, and HEMs. This application is able to measure total changes across the whole retinal image. Hence, retinal images that belong to the same patients are examined in order to monitor the trend of the illness. To make a comparison with other methods, a Naïve Bayes method is applied for segmentation of DR. The performance of the system, tested on different data sets including various qualities of retinal fundus images, is over 95% in detection of the optic disc (OD), and 90% in segmentation of the DR.

Anisotropy-based robust focus measure for non-mydriatic retinal imaging

Non-mydriatic retinal imaging is an important tool for diagnosis and progression assessment of ophthalmic diseases. Because it does not require pharmacological dilation of the patient's pupil, it is essential for screening programs performed by non-medical personnel. A typical camera is equipped with a manual focusing mechanism to compensate for the refractive errors in the eye. However, manual focusing is error prone, especially when performed by inexperienced photographers. In this work, we propose a new and robust focus measure based on a calculation of image anisotropy which, in turn, is evaluated from the directional variance of the normalized discrete cosine transform. Simulation and experimental results demonstrate the effectiveness of the proposed focus measure.

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

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

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.

Automated segmentation of intraretinal cystoid fluid in optical coherence tomography

Cystoid macular edema (CME) is observed in a variety of ocular disorders and is strongly associated with vision loss. Optical coherence tomography (OCT) provides excellent visualization of cystoid fluid, and can assist clinicians in monitoring the progression of CME. Quantitative tools for assessing CME may lead to better metrics for choosing treatment protocols. To address this need, this paper presents a fully automated retinal cyst segmentation technique for OCT image stacks acquired from a commercial scanner. The proposed method includes a computationally fast bilateral filter for speckle denoising while maintaining CME boundaries. The proposed technique was evaluated in images from 16 patients with vitreoretinal disease and three controls. The average sensitivity and specificity for the classification of cystoid regions in CME patients were found to be 91% and 96%, respectively, and the retinal volume occupied by cystoid fluid obtained by the algorithm was found to be accurate within a mean and median volume fraction of 1.9% and 0.8%, respectively.

Coupled parametric model for estimation of visual field tests based on OCT macular thickness maps, and vice versa, in glaucoma care

The current standard of care for glaucoma patients consists of functional assessment of vision via visual field (VF) testing, which is sensitive but subjective, time-consuming, and often unreliable. A new imaging technology, Fourier domain optical coherence tomography (OCT), is being introduced to assess the structural characteristics of the macula. This new complementary exam is efficient, objective, and reliable. A complex, but consistent, relationship exists between the structural information provided by macular OCT and the functional information gathered by VF maps. We propose a learning-based framework with the ability to estimate the VF map based on OCT macular thickness measurements as input (and vice versa). Central to this algorithmic framework is a coupled parametric model that captures not only the individual variabilities of the OCT macular thickness measurements and the VF maps, but also their co-dependencies. This model is derived by applying principal component analysis (PCA) to a library consisting of various pairs of OCT and VF maps. The parameters of this coupled model are obtained by solving a linear least squares problem. Next, these estimated parameters are used, in conjunction with the eigenvectors derived from PCA, to compute the estimate. The accuracy of this coupled parametric estimation model was evaluated by performing multiple leave-one-out cross validation experiments.

Exudate-based diabetic macular edema detection in fundus images using publicly available datasets

Diabetic macular edema (DME) is a common vision threatening complication of diabetic retinopathy. In a large scale screening environment DME can be assessed by detecting exudates (a type of bright lesions) in fundus images. In this work, we introduce a new methodology for diagnosis of DME using a novel set of features based on colour, wavelet decomposition and automatic lesion segmentation. These features are employed to train a classifier able to automatically diagnose DME through the presence of exudation. We present a new publicly available dataset with ground-truth data containing 169 patients from various ethnic groups and levels of DME. This and other two publicly available datasets are employed to evaluate our algorithm. We are able to achieve diagnosis performance comparable to retina experts on the MESSIDOR (an independently labelled dataset with 1200 images) with cross-dataset testing (e.g., the classifier was trained on an independent dataset and tested on MESSIDOR). Our algorithm obtained an AUC between 0.88 and 0.94 depending on the dataset/features used. Additionally, it does not need ground truth at lesion level to reject false positives and is computationally efficient, as it generates a diagnosis on an average of 4.4s (9.3s, considering the optic nerve localisation) per image on an 2.6 GHz platform with an unoptimised Matlab implementation.

In vivo chromium-enhanced MRI of the retina

Chromium (Cr) has been used histologically to stabilize lipid fractions in the retina and is suggested to enhance oxidizable lipids in brain MRI. This study explored the feasibility, sensitivity, and specificity of in vivo chromium-enhanced MRI of retinal lipids by determining its spatiotemporal profiles and toxic effect after intravitreal Cr(VI) injection to normal adult rats. One day after 3 μL Cr(VI) administration at 1-100 mM, the retina exhibited a dose-dependent increase in T1-weighted hyperintensity until 50 mM. Time-dependently, significant T1-weighted hyperintensity persisted up to 2 weeks after 10 mM Cr(VI) administration. Three-dimensional chromium-enhanced MRI of ex vivo normal eyes at isotropic 50-μm resolution showed at least five alternating bands across retinal layers, with the outermost layer being the brightest. This agreed with histology indicating alternating lipid contents with the highest level in the photoreceptor layer of the outer retina. Although Cr(VI) reduction may induce oxidative stress and depolymerize microtubules, manganese-enhanced MRI after chromium-enhanced MRI showed a dose-dependent effect of Cr toxicity on manganese uptake and axonal transport along the visual pathway. These results potentiated future longitudinal chromium-enhanced MRI studies on retinal lipid metabolism upon further optimization of Cr doses with visual cell viability.