Detection of red reflex asymmetry by pediatric residents using the Brückner reflex versus the MTI photoscreener

Automatic Refractive Error Estimation Using Deep Learning-Based Analysis of Red Reflex Images

Purpose/Background: We evaluate how a deep learning model can be applied to extract refractive error metrics from pupillary red reflex images taken by a low-cost handheld fundus camera. This could potentially provide a rapid and economical vision-screening method, allowing for early intervention to prevent myopic progression and reduce the socioeconomic burden associated with vision impairment in the later stages of life. Methods: Infrared and color images of pupillary crescents were extracted from eccentric photorefraction images of participants from Choithram Hospital in India and Dargaville Medical Center in New Zealand. The pre-processed images were then used to train different convolutional neural networks to predict refractive error in terms of spherical power and cylindrical power metrics. Results: The best-performing trained model achieved an overall accuracy of 75% for predicting spherical power using infrared images and a multiclass classifier. Conclusions: Even though the model's performance is not superior, the proposed method showed good usability of using red reflex images in estimating refractive error. Such an approach has never been experimented with before and can help guide researchers, especially when the future of eye care is moving towards highly portable and smartphone-based devices.

Retinoblastoma: Review and new insights

Retinoblastoma (Rb), the most frequent malignant intraocular tumor in childhood, is caused by mutations in the retinoblastoma gene (RB1) situated on chromosome 13q14.2. The incidence of retinoblastoma is approximately 1 in 17,000 live births with approximately 8,000 new cases diagnosed each year worldwide. Rb is the prototypical hereditary cancer in humans. Autosomal dominant inheritance is seen in 30-40% of cases whereas the non-inherited sporadic type accounts for the remaining 60-70%. Rb arises due to inactivation of both alleles of the Rb tumor suppressor gene, which results in a defective Rb protein (pRB) with subsequent cell cycle impairment and uncontrolled cell proliferation. Patients with Rb have survival rates higher than 95-98% in industrialized countries but mortality remains high in developing countries. For example, the mortality rate in Africa is 70%. In all cases of intraocular and extraocular retinoblastoma, there is a need for new therapies that are more effective and carry less risk of toxicity. The Bruckner test is a practical and easy test for the detection of Rb, this test consists of assessing the fundus reflex through the pupil (red reflex) in both eyes simultaneously with a bright coaxial light produced with the direct ophthalmoscope. Rb can be detected by the Bruckner test showing a pupil that shines white or “Leukocoria”. Although the diagnosis of Rb remains essentially clinical, the newly identified biomarkers could contribute to early molecular detection, timely detection of micrometastases and establish new therapeutic options for Rb.

Comparison between wide-field digital imaging system and the red reflex test for universal newborn eye screening in Brazil

PURPOSE

To compare neonatal eye screening using the red reflex test (RRT) versus the wide-field digital imaging (WFDI) system.

METHODS

Prospective cohort study. Newborns (n = 380, 760 eyes) in the Maternity Ward of Irmandade Santa Casa de Misericórdia de São Paulo hospital from May to July 2014 underwent RRT by a paediatrician and WFDI performed by the authors. Wide-field digital imaging (WFDI) images were analysed by the authors. Validity of the paediatrician's RRT was assessed by unweighted kappa [κ] statistic, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).

RESULTS

While WFDI showed abnormalities in 130 eyes (17.1%), RRT was only abnormal in 13 eyes (1.7%). Wide-field digital imaging (WFDI) detected treatable retina pathology that RRT missed including hyphema, CMV retinitis, FEVR and a vitreous haemorrhage. The sensitivity of the paediatrician's RRT to detect abnormalities was poor at 0.77% (95% confidence interval, CI, 0.02%-4.21%) with a PPV of only 7.69% (95% CI, 1.08%-38.85%). Overall, there was no agreement between screening modalities (κ = -0.02, 95% CI, -0.05 to 0.01). The number needed to screen to detect ocular abnormalities using WFDI was 5.9 newborns and to detect treatable abnormalities was 76 newborns.

CONCLUSION

While RRT detects gross abnormalities that preclude visualization of the retina (i.e. media opacities and very large tumours), only WFDI consistently detects subtle treatable retina and optic nerve pathology. With a higher sensitivity than the current gold standard, universal WFDI allows for early detection and management of potentially blinding ophthalmic disease missed by RRT.

Ophthalmology examination during well-child visits in primary care health centres: Knowledge and difficulties experienced by health workers

INTRODUCTION

The red reflex examination (RRE) and visual acuity testing (VA) is a mandatory part of the examination during the well-child visits (WCV) in primary health care centres of the public system of health in Chile. The eye examination is aimed at the early detection of severe eye diseases in children, such as retinoblastoma, congenital cataracts, and amblyopia. The knowledge and difficulties experienced by health workers in primary care health centres for evaluating the red reflex during WCV in Chile is unknown.

MATERIAL AND METHODS

A survey was performed in primary community health centres of XXX Santiago de Chile.

RESULTS

The WCV were mainly performed by physicians (45.2%) and nurses (35.8%). Only 34% of health workers performed the red reflex test, and 42.3% checked VA during the WCV. The main reasons for not doing it include the lack of direct ophthalmoscopes and VA charts (55.2% and 43.9%, respectively) at their centres, and not having the knowledge or skills (29.3% and 22%, respectively) to properly perform these clinical tests.

CONCLUSION

In this series, the eye examination of children attending WCV was unfrequently performed. A better implementation of the health centres and training of the health workers are needed in order to improve the access and quality of the paediatric eye examination in primary health care institutions in Chile.

Diagnostic test accuracy of Spot and Plusoptix photoscreeners in detecting amblyogenic risk factors in children: a systemic review and meta-analysis

PURPOSE

Amblyopia is a permanent visual impairment developed in early childhood. Recently, instrument-based photoscreeners have been widely used for vision screening to detect amblyopia risk factors, which is key for the prevention and treatment of amblyopia. This meta-analysis aims to evaluate the diagnostic test accuracy of Spot and Plusoptix photoscreeners in detecting risk factors for amblyopia in children.

METHODS

An electronic literature search was performed in October 2018 using the MEDLINE, Embase, and Web of Science databases. Twenty-one publications with a total of 5022 subjects were included. Subjects given a comprehensive examination were considered to have amblyopia or amblyogenic risk factors based on the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) referral criteria guidelines.

RESULTS

The overall sensitivity for Spot and Plusoptix was 87.7% and 89.4%, respectively (p = 0.38); the specificity was 78.0% and 89.9%, respectively (p = 0.90). For subjects under 7 years old (preschool children), the overall sensitivity for Spot and Plusoptix was 91.7% and 90.2%, respectively (p = 0.81); the specificity was 82.6% and 93.0%, respectively (p = 0.46).

CONCLUSION

Both Spot and Plusoptix photoscreeners demonstrated good sensitivity and specificity in detecting amblyopia risk factors in children, especially at preschool ages. There was no significant difference in diagnostic test accuracy between them.

Can the Red Reflex Test Detect Unequal Refractive Error?

OBJECTIVE

To determine the accuracy of the red reflex test in the detection of anisometropia.

STUDY DESIGN

This prospective, single-masked study enrolled new patients younger than the age of 18 years who had not undergone pharmacologic pupillary dilation. A fellow who was masked to all clinical information illuminated both eyes with a direct ophthalmoscope in a darkened room from a distance of 1 m, assessing whether the red reflex between the 2 eyes was symmetric or asymmetric. The patient was then dilated, and cylcoplegic refraction was performed by an attending pediatric ophthalmologist. Exclusion criteria included the presence of strabismus, anisocoria, previous intraocular surgery, media opacity, leukocoria, or nystagmus. Sensitivity was compared with a null hypothesized value of 50% using a 1-sided binomial test.

RESULTS

Ninety-two patients with a mean age of 7.3 years (range 3 months to 16 years) were enrolled. With spherical anisometropia greater than or equal to 0.125 diopters, the sensitivity of the red reflex test was 90.6% and the specificity was 58.3%. With cylindrical anisometropia greater than or equal to 0.25 diopters, the sensitivity of the red reflex test was 81.3% and the specificity was 70%. Anisometropia greater than 1.5 diopters in spherical equivalent (4 patients, range -10.625 to -2.625) or cylinder (3 patients, range 1.75-2.25) was accurately detected by red reflex testing in each case.

CONCLUSIONS

The red reflex test can be an accurate screening tool to detect anisometropia when performed by an ophthalmologist.

Tests for detecting strabismus in children aged 1 to 6 years in the community

BACKGROUND

Strabismus (misalignment of the eyes) is a risk factor for impaired visual development both of visual acuity and of stereopsis. Detection of strabismus in the community by non-expert examiners may be performed using a number of different index tests that include direct measures of misalignment (corneal or fundus reflex tests), or indirect measures such as stereopsis and visual acuity. The reference test to detect strabismus by trained professionals is the cover‒uncover test.

OBJECTIVES

To assess and compare the accuracy of tests, alone or in combination, for detection of strabismus in children aged 1 to 6 years, in a community setting by non-expert screeners or primary care professionals to inform healthcare commissioners setting up childhood screening programmes.Secondary objectives were to investigate sources of heterogeneity of diagnostic accuracy.

SEARCH METHODS

We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2016, Issue 12) (which contains the Cochrane Eyes and Vision Trials Register) in the Cochrane Library, the Health Technology Assessment Database (HTAD) in the Cochrane Library (2016, Issue 4), MEDLINE Ovid (1946 to 5 January 2017), Embase Ovid (1947 to 5 January 2017), CINAHL (January 1937 to 5 January 2017), Web of Science Conference Proceedings Citation Index-Science (CPCI-S) (January 1990 to 5 January 2017), BIOSIS Previews (January 1969 to 5 January 2017), MEDION (to 18 August 2014), the Aggressive Research Intelligence Facility database (ARIF) (to 5 January 2017), the ISRCTN registry (www.isrctn.com/editAdvancedSearch); searched 5 January 2017, ClinicalTrials.gov (www.clinicaltrials.gov); searched 5 January 2017 and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en); searched 5 January 2017. We did not use any date or language restrictions in the electronic searches for trials. In addition, orthoptic journals and conference proceedings without electronic listings were searched.

SELECTION CRITERIA

All prospective or retrospective population-based test accuracy studies of consecutive participants were included. Studies compared a single or combination of index tests with the reference test. Only those studies with sufficient data for analysis were included specifically to calculate sensitivity and specificity and determine diagnostic accuracy.Participants were aged 1 to 6 years. Studies reporting participants outside this range were included if subgroup data were available.Permitted settings included population-based vision screening programmes or opportunistic screening programmes, such as those performed in schools.

DATA COLLECTION AND ANALYSIS

We used standard methodological procedures expected by Cochrane. In brief, two review authors independently assessed titles and abstracts for eligibility and extracted the data, with a third senior author resolving any disagreement. We analysed data primarily for specificity and sensitivity.

MAIN RESULTS

One study from a total of 1236 papers, abstracts and trials was eligible for inclusion with a total number of participants of 335 of which 271 completed both the screening test and the gold standard test. The screening test using an automated photoscreener had a sensitivity of 0.46 (95% confidence interval (CI) 0.19 to 0.75) and specificity of 0.97 (CI 0.94 to 0.99). The overall number affected by strabismus was low at 13 (4.8%).

AUTHORS' CONCLUSIONS

There is very limited data in the literature to ascertain the accuracy of tests for detecting strabismus in the community as performed by non-expert screeners. A large prospective study to compare methods would be required to determine which tests have the greatest accuracy.

Sensitivity and Specificity of Red Reflex Test in Newborn Eye Screening

OBJECTIVES

To validate the sensitivity of an isolated red reflex test in detection of ocular abnormalities of the anterior and posterior segments in newborns.

STUDY DESIGN

Red reflex test and comprehensive eye examinations including external inspection, pupil examination, hand-held slit lamp examination, and RetCam fundus imaging (Clarity Medical Systems, Pleasanton, California) were performed in 7641 newborns. All results were documented as negative or positive. Sensitivity and specificity of red reflex test were calculated by the use of comprehensive eye examinations as the reference standard. Anterior abnormalities were separated from posterior abnormalities, and the sensitivity of red reflex test for each group was calculated.

RESULTS

The proportion of abnormalities that were correctly classified by red reflex test was greater in anterior segment group (sensitivity = 99.6%, 95% CI 97.1%-100%) than in the posterior group (sensitivity = 4.1%, 95% CI 3.3%-5.1%, χ² = 1521.382, φ = 0.836, P < .001).

CONCLUSIONS

The red reflex test was a useful universal screening tool in detection of anterior abnormalities; however, the test has limitations in detection of posterior abnormalities. The generalization of these results to infants and children and observers with varying levels of expertise may need to be established further.

Advantages, limitations, and diagnostic accuracy of photoscreeners in early detection of amblyopia: a review

Amblyopia detection is important to ensure proper visual development and avoid permanent decrease of visual acuity. This condition does not produce symptoms, so it is difficult to diagnose if a vision problem actually exists. However, because amblyopia treatment is limited by age, early diagnosis is of paramount relevance. Traditional vision screening (conducted in <3 years) is related with difficulty in getting cooperation from a subject to conduct the eye exam, so accurate objective methods to improve amblyopia detection are necessary. Handheld devices used for photoscreening or autorefraction could offer advantages to improve amblyopia screening because they reduce exploration time to just few seconds, no subject collaboration is needed, and they provide objective information. The purpose of this review is to summarize the main functions and clinical applicability of commercially available devices for early detection of amblyopia and to describe their differences, advantages, and limitations. Although the studies reviewed are heterogeneous (due to wide differences in referral criteria, use of different risk factors, different types of samples studied, etc), these devices provide objective measures in a quick and objective way with a simple outcome report: retest, pass, or refer. However, due to major limitations, these devices are not recommended, and their use in clinical practice is limited.

Assessment of a modification of Brückner's test as a screening modality for anisometropia and strabismus

BACKGROUND

Current amblyopia screening methods are not cost effective.

AIM

To evaluate the diagnostic capability of a modified Brückner test (MBT) for amblyopiogenic risk factors.

MATERIALS AND METHODS

We applied the MBT using the streak retinoscope to identify anisometropia and strabismus by noting an inter-ocular difference in movement and glow, from children who failed 6/9 Snellen on community vision screening, followed by comprehensive eye examination. STATISITICS: Data were analyzed by 2 × 2 tables for diagnostic test parameters (95% CI).

RESULTS

From 7998 children vision-screened, 392 failed 6/9 VA and were referred. Since 34 failed to reach the centers, and 15 were excluded due to poor/ no glow, data from 343 was analyzed. The prevalence of anisometropia of 0.5D was 17%, of 1D was 11% and of strabismus 5%. For the MBT the accuracy was ≥ 90% (95%CI 89% to 97%) over the three outcomes. The sensitivity, specificity, NPV and +LR for anisometropia of 0.5D were: 0.57 (0.48, 0.64), 0.97 (0.95, 0.98), 0.92 (0.90, 0.93) and 18 (9.7, 35); for 1D: 0.74(0.60, 0.82), 0.95 (0.94, 0.97), 0.97 (0.95, 0.98) and 16 (9.3, 28); and for strabismus: 0.5 (0.32, 0.66), 0.98 (0.97, 0.98), 0.97 (0.96, 0.98) and 20 (9.1, 42).

CONCLUSION

Our data suggests that the MBT is highly accurate and useful for ruling in anisometropia and strabismus in children who fail 6/9 Snellen. The MBT needs further validation, both by different care givers and on differing populations. It offers an affordable, portable, and clinically useful tool to detect anisometropia and strabismus. We suggest that performing an MBT prior to uniocular retinosocpy should be a routine practice.

The Brückner test: extended distance improves sensitivity for ametropia

BACKGROUND

The Brückner test is usually performed at a distance of 1 m. Due to optical reasons, the test should detect ametropia more sensitively at an extended distance. We compared the sensitivity of the test to detect unilateral ametropia at a distance of 4 m versus the traditional distance of 1 m.

PATIENTS AND METHODS

In this study, five blinded experienced observers (experts) performed the Brückner transillumination test on five emmetropic subjects (age 22.3 +/- 2.3 years) at a distance of 1 m which was then extended up to 4 m. Unilateral ametropia was simulated by both spherical and cylindrical lenses of -1, -2, -3, -4, and +1, +2, +3, +4 diopters in front of one eye and a plano lens in front of the other eye in a randomised order. Controls with plano lenses in front of both eyes were interspersed. The test was considered positive in case of any difference between the red reflexes of both eyes. For spherical ametropia, the procedure was repeated with 25 medical students who examined one subject each.

RESULTS

At a distance of 1 m, unilateral myopia/hypermetropia of more than 1 diopter was detected in 78.7%/71.3% by experts and in 34.7%/53.3% by students. At 4 m, detection rates increased to 99.3%/96.0% and 98.7%/100% respectively. Rates of false positive findings for experts vs students were 3.1% vs 1.5% at 1 m and 4.0% vs 3.0% at 4 m. For unilateral astigmatism, experts' detection rates were similar at 1 m (64.7%/70.0%) and lower at 4 m (91.7%/90.3%), depending on the astigmatism axis.

CONCLUSIONS

The sensitivity of the Brückner reflex for anisometropia improves by extension of the examination distance, especially in the hands of less experienced observers. To detect ametropia more sensitively, a test distance of 4 m is recommended.

A new visual evoked potential system for vision screening in infants and young children

INTRODUCTION

With a prevalence of 3-5%, amblyopia represents a major public health problem. Effective treatment depends on early detection, and a broad consensus of professional opinion supports vision screening of infants and young children. No single method of screening has been demonstrated to be superior in detecting amblyopia and all methods have significant limitations.

METHODS

We assessed a new, "child-friendly" visual evoked potential (VEP) system (ENFANTtrade mark II, Diopsys Corp., Metuchen, NJ) for use in screening. We studied 122 children, aged 6 months to 5 years, comparing test results in a masked fashion to results of standard ophthalmologic examinations. A statistical program analyzed VEP differences between fellow eyes to determine a "pass" or "fail" for each child. For verbal patients, clinical amblyopia was defined as an interocular difference of two or more lines in best-corrected visual acuity. For preverbal patients, clinical amblyopia was defined by the clinician's decision to treat with occlusion or atropine penalization. Preverbal children with significant refractive errors or structural eye pathology were also considered clinically abnormal.

RESULTS

The test was completed by 94% of the study group, each child requiring an average of 10 minutes to complete testing of both eyes. The sensitivity was 0.973, the specificity 0.808, the positive predictive value 0.706, and the negative predictive value 0.984.

CONCLUSION

With its easy electrode placement and rapid, attractive stimulus, the new system overcomes technical difficulties which were associated with older VEP techniques. The test shows promise as a screening tool for detecting amblyopia and other visual deficits in young children.