Artificial Intelligence in Cataract Surgery: A Systematic Review.

Purpose: The purpose of this study was to assess the current use and reliability of artificial intelligence (AI)-based algorithms for analyzing cataract surgery videos. Methods: A systematic review of the literature about intra-operative analysis of cataract surgery videos with machine learning techniques was performed. Cataract diagnosis and detection algorithms were excluded. Resulting algorithms were compared, descriptively analyzed, and metrics summarized or visually reported. The reproducibility and reliability of the methods and results were assessed using a modified version of the Medical Image Computing and Computer-Assisted (MICCAI) checklist. Results: Thirty-eight of the 550 screened studies were included, 20 addressed the challenge of instrument detection or tracking, 9 focused on phase discrimination, and 8 predicted skill and complications. Instrument detection achieves an area under the receiver operator characteristic curve (ROC AUC) between 0.976 and 0.998, instrument tracking an mAP between 0.685 and 0.929, phase recognition an ROC AUC between 0.773 and 0.990, and complications or surgical skill performs with an ROC AUC between 0.570 and 0.970. Conclusions: The studies showed a wide variation in quality and pose a challenge regarding replication due to a small number of public datasets (none for manual small incision cataract surgery) and seldom published source code. There is no standard for reported outcome metrics and validation of the models on external datasets is rare making comparisons difficult. The data suggests that tracking of instruments and phase detection work well but surgical skill and complication recognition remains a challenge for deep learning. Translational Relevance: This overview of cataract surgery analysis with AI models provides translational value for improving training of the clinician by identifying successes and challenges.

The impact of COVID-19 on compliance to amblyopia treatment in a tertiary eye care center.

Purpose: To investigate the impact of the coronavirus disease 2019 (COVID-19) outbreak on compliance to amblyopia therapy from the perspective of parents of amblyopic children. Methods: The participants for the qualitative study were enrolled from the medical records of a tertiary eye care center during the COVID-19 pandemic. The trained researcher conducted telephonic interviews consisting of validated 15 open-ended questions for 15 min. The questions were related to patients' compliance toward amblyopia treatment and the date of follow-up with their practitioners. The data collected were entered on Excel sheets in the participant's own words and later converted to a transcript for analysis. Results: A total of 217 parents of children with amblyopia who were due for follow-up were contacted through telephone. The response rate for willingness to participate was only 36% (n = 78). Of these, 76% (n = 59) of parents reported that their child was compliant toward the treatment during the therapy period and 69% reported that their child was currently not on treatment for amblyopia. Conclusion: In the current study, we have observed that although reported compliance by parents during the therapy period was good, around 69% of patients had discontinued amblyopia therapy. The main reason for the discontinuation of therapy was the failure to visit the hospital during the scheduled follow-up with the eye care practitioner.

Validation of a simple-to-use, affordable, portable, wavefront aberrometry-based auto refractometer in the adult population: A prospective study.

BACKGROUND: Refraction is one of the key components of a comprehensive eye examination. Auto refractometers that are reliable and affordable can be beneficial, especially in a low-resource community setting. The study aimed to validate the accuracy of a novel wave-front aberrometry-based auto refractometer, Instaref R20 against the open-field system and subjective refraction in an adult population. METHODS: All the participants underwent a comprehensive eye examination including objective refraction, subjective acceptance, anterior and posterior segment evaluation. Refraction was performed without cycloplegia using WAM5500 open-field auto refractometer (OFAR) and Instaref R20, the study device. Agreement between both methods was evaluated using Bland-Altman analysis. The repeatability of the device based on three measurements in a subgroup of 40 adults was assessed. RESULTS: The refractive error was measured in 132 participants (mean age,30.53 ± 9.36 years, 58.3% female). The paired mean difference of the refraction values of the study device against OFAR was - 0.13D for M, - 0.0002D (J0) and - 0.13D (J45) and against subjective refraction (SR) was - 0.09D (M), 0.06 (J0) and 0.03D (J45). The device agreed within +/- 0.50D of OFAR in 78% of eyes for M, 79% for J0 and 78% for J45. The device agreed within +/- 0.5D of SR values for M (84%), J0 (86%) and J45 (89%). CONCLUSION: This study found a good agreement between the measurements obtained with the portable autorefractor against open-field refractometer and SR values. It has a potential application in population-based community vision screening programs for refractive error correction without the need for highly trained personnel.

Validation of a Simple-to-Use, Affordable, Portable Wavefront Aberrometry-Based Auto Refractometer in a Paediatric Population.

Purpose: InstaRef R20 is a handheld, affordable auto refractometer based on Shack Hartmann aberrometry technology. The study's objective was to compare InstaRef R20's performance for identifying refractive error in a paediatric population to that of standard subjective and objective refraction under both pre- and post-cycloplegic conditions. Methods: Refraction was performed using 1) standard clinical procedure consisting of retinoscopy followed by subjective refraction (SR) under pre- and post-cycloplegic conditions and 2) InstaRef R20. Agreement between both methods was evaluated using Bland-Altman analysis. The repeatability of the device based on three measurements in a subgroup of 20 children was assessed. Results: The refractive error was measured in 132 children (mean age 12.31 ± 3 years). The spherical equivalent (M) and cylindrical components (J0 and J45) of the device had clinically acceptable differences (within ±0.50D) and acceptable agreement compared to standard pre- and post-cycloplegic manual retinoscopy and subjective refraction (SR). The device agreed within ± 0.50D of retinoscopy in 67% of eyes for M, 78% for J0 and 80% for J45 and within ± 0.50D of SR in 70% for M and 77% for cylindrical components. Conclusion: InstaRef R20 has an acceptable agreement compared to standard retinoscopy in paediatric population. The measurements from this device can be used as a starting point for subjective acceptance. The device being simple to use, portable, reliable and affordable has the potential for large-scale community-based refractive error detection.

Binocular therapy as primary intervention in adults with anisometropic amblyopia.

PURPOSE: Refractive correction and patching is the timetested mainstay of treatment for anisometropic amblyopia within the critical period of visual development. Binocular therapies using dichoptic training which overcome suppression by balancing the contrast between two eyes has been increasingly gaining ground. We evaluated the efficacy of dichoptic training in the adult population with anisometropic amblyopia. This study aims to evaluate the effectiveness of dichopticbased active vision therapy, using "VisuoPrime" software as primary intervention, in adults with anisometropic amblyopes. MATERIALS AND METHODS: A prospective interventional study in adults (18-40 years) with anisometropic amblyopia was conducted from August 2019 to March 2020. METHODS: Twentynine subjects with anisometropic amblyopia played binocular games through "VisuoPrime" software 30 min daily for 6 weeks. Bestcorrected visual acuity (BCVA) and binocularity was assessed at 1 and 3 months. Student's paired ttest, Wilcoxon signedrank sum test and MannWhitney tests were used. Statistical package of SPSS version 20.0 was used for analysis, considering P < 0.05 as statistically significant. RESULTS: BCVA of the amblyopic eye improved from 0.60 ± 0.40 logMAR to 0.45 ± 0.29 logMAR and 0.38 ± 0.23 logMAR at 1 and 3 months, respectively (P = 0.0001). Near acuity improved from 0.21 ± 0.14 to 0.14 ± 0.08 logMAR and 0.1 ± 0.04 logMAR at 1 and 3 months respectively (P < 0.0001). Improvement in stereopsis was observed in 24% of subjects which maintained at 3 month followup. CONCLUSION: Dichopticbased active vision therapy using "VisuoPrime" software was effective as a primary modality in adults with anisometropic amblyopia.

Keratoconus Classifier for Smartphone-based Corneal Topographer.

Keratoconus is a severe eye disease that leads to deformation of the cornea. It impacts people aged 10-25 years and is the leading cause of blindness in that demography. Corneal topography is the gold standard for keratoconus diag-nosis. It is a non-invasive process performed using expensive and bulky medical devices called corneal topographers. This makes it inaccessible to large populations, especially in the Global South. Low-cost smartphone-based corneal topographers, such as SmartKC, have been proposed to make keratoconus diagnosis accessible. Similar to medical-grade topographers, SmartKC outputs curvature heatmaps and quantitative metrics that need to be evaluated by doctors for keratoconus diagnosis. An auto-matic scheme for evaluation of these heatmaps and quantitative values can play a crucial role in screening keratoconus in areas where doctors are not available. In this work, we propose a dual-head convolutional neural network (CNN) for classifying keratoconus on the heatmaps generated by SmartKC. Since SmartKC is a new device and only had a small dataset (114 sam-ples), we developed a 2-stage transfer learning strategy-using historical data collected from a medical-grade topographer and a subset of SmartKC data-to satisfactorily train our network. This, combined with our domain-specific data augmentations, achieved a sensitivity of 91.3% and a specificity of 94.2%.

Cost-Effectiveness of photoscreeners in screening at-risk amblyopia in Indian children.

Objectives: Amblyopia is the most common cause of preventable vision impairment in children. This study tried to evaluate the cost-effectiveness of photoscreeners in identifying refractive errors making children at risk of amblyopia. Materials and Methods: This was designed as a prospective, open-label, randomized controlled study to compare the cost-effectiveness of photoscreeners (2WIN Adaptica and Plusoptix) versus autorefractor and Mohindra retinoscopy to identify the at-risk amblyopia. This study was conducted from October 2018 to December 2018 in coordination with Nanna Kannu school screening project and children in the age group of 5-11 years were included in the study. The photoscreening was done by lay screeners. A supervising ophthalmologist collated all data. Cost-effectiveness was calculated for the individual instruments and compared. The average time is taken for each instrument to record the refractive error was calculated and compared with the Mohindra retinoscopy. Results: Number of children included in this study was 2910. The mean age of the children was 7.82 + 0.65 years. The mean time taken for 2WIN Adaptica was 6 sec, Topcon -10 s, and Plusoptix -4 s. Plusoptix showed the minimal time for screening a child when compared to all other methods. The Plusoptix and 2WIN Adaptica were shown to be cost-effective when compared to Mohindra retinoscopy (P < 0.05). Conclusion: Photoscreening using Plusoptix and 2WIN Adaptica with lay screeners was < seven times the cost incurred and five times faster than the trained optometrists using either autorefractor or Mohindra retinoscopy and thus more cost-effective in screening a large number of young children with at-risk amblyopia at the community level.

A questionnaire-based assessment of Safe Eye Examination (SEE) technique.

PURPOSE: This study aims to assess the novel and innovative method of Safe Eye Examination (SEE) technique using the model eye for the purpose of teaching, training, and resident examination. METHODS: A questionnaire-based study (Descriptive Data) with 53 participants, including ophthalmology residents, fellows in various subspecialties, and trainee optometrists was used. In our study, we used the Reti Eye model, but instead of the usual retina template sheet, we used prominent pathological fundus photographs loaded into the model eye. The study participants were asked to view prominent pathological fundus images printed on a matte finish paper. A circular image was cut and was placed in the Reti Eye model. The candidates were made to perform indirect ophthalmoscopy with a + 20D lens and to fill up a Google image assessment scale questionnaire with characteristics, including pixelation, sharpness, contrast, reflexes, blotchy appearance, and diagnostic confidence, which were then analyzed and depicted. Association between categorical variables was analyzed using Fisher exact test and Chi-square test. A P value of less than 0.05 was considered statistically significant. All data were analyzed with a statistical software package (SPSS, Version 16.0 for Windows). RESULTS: The number of positive responses (>90%) obtained for the pixelation, sharpness, contrast, reflexes, blotchy appearance, and diagnostic confidence of the image viewed were statistically more significant than the negative responses (P < 0.05). CONCLUSION: The SEE technique of using the model eye can be incorporated for teaching, training, and skill assessment in the examinations in these difficult times of COVID-19 (coronavirus disease 2019) pandemic.

Effectiveness of Kanna photoscreener in detecting amblyopia risk factors.

Purpose: Amblyopia is a significant public health problem. Photoscreeners have been shown to have significant potential for screening; however, most are limited by cost and display low accuracy. The purpose of this study was validate a novel artificial intelligence (AI) and machine learning-based facial photoscreener "Kanna," and to determine its effectiveness in detecting amblyopia risk factors. Methods: A prospective study that included 654 patients aged below 18 years was conducted in our outpatient clinic. Using an android smartphone, three images of each the participants' face were captured by trained optometrists in dark and ambient light conditions and uploaded onto Kanna. Deep learning was used to create an amblyopia risk score based on our previous study. The algorithm generates a risk dashboard consisting of six values: five normalized risk scores for ptosis, strabismus, hyperopia, myopia and media opacities; and one binary value denoting if a child is "at-risk" or "not at-risk." The presence of amblyopia risk factors (ARF) as determined on the ophthalmic examination was compared with the Kanna photoscreener. Results: Correlated patient data for 654 participants were analyzed. The mean age of the study population was 7.87 years. The algorithm had an F-score, 85.9%; accuracy, 90.8%; sensitivity, 83.6%; specificity, 94.5%; positive predictive value, 88.4%; and negative predictive value, 91.9% in identifying amblyopia risk factors. The P value for the amblyopia risk calculation was 8.5 × 10-142 implying strong statistical significance. Conclusion: The Kanna photo-based screener that uses deep learning to analyze photographs is an effective alternative for screening children for amblyopia risk factors.

Application of deep learning and image processing analysis of photographs for amblyopia screening.

Purpose: Photo screeners and autorefractors have been used to screen children for amblyopia risk factors (ARF) but are limited by cost and efficacy. We looked for a deep learning and image processing analysis-based system to screen for ARF. Methods: An android smartphone was used to capture images using a specially coded application that modified the camera setting. An algorithm was developed to process images taken in different light conditions in an automated manner to predict the presence of ARF. Deep learning and image processing models were used to segment images of the face. Light settings and distances were tested to obtain the necessary features. Deep learning was thereafter used to formulate normalized risks using sigmoidal models for each ARF creating a risk dashboard. The model was tested on 54 young adults and results statistically analyzed. Results: A combination of low-light and ambient-light images was needed for screening for exclusive ARF. The algorithm had an F-Score of 73.2% with an accuracy of 79.6%, a sensitivity of 88.2%, and a specificity of 75.6% in detecting the ARF. Conclusion: Deep-learning and image-processing analysis of photographs acquired from a smartphone are useful in screening for ARF in children and young adults for a referral to doctors for further diagnosis and treatment.

Diabetic Retinopathy Screening Using Smartphone-Based Fundus Imaging in India.

PURPOSE: Early detection and treatment can prevent irreversible blindness from diabetic retinopathy (DR), which is the leading cause of visual impairment among working-aged adults worldwide. Some 80% of affected persons live in low- and middle-income countries, yet lack of resources has largely prevented DR screening implementation in these world regions. Smartphone-based fundus imaging (SBFI) allows for low-cost mobile fundus examination using an adapter on a smartphone; however, key aspects such as image quality, diagnostic accuracy, and comparability of different approaches have not been systematically assessed to date. DESIGN: Evaluation of diagnostic technology. PARTICIPANTS: A total of 381 eyes of 193 patients with diabetes were recruited at outreach eye clinics in South India. METHODS: We compared 4 technically different approaches of SBFI (3 approaches based on direct and 1 approach based on indirect ophthalmoscopy) in terms of image quality and diagnostic accuracy for DR screening. MAIN OUTCOME MEASURES: Image quality (sharpness/focus, reflex artifacts, contrast, and illumination), field-of-view, examination time, and diagnostic accuracy for DR screening were analyzed against conventional fundus photography and clinical examination. RESULTS: Smartphone-based fundus imaging based on indirect ophthalmoscopy yielded the best image quality (P < 0.01), the largest field-of-view, and the longest examination time (111 vs. 68-86 seconds, P < 0.0001). Agreement with the reference standard (Cohen's kappa 0.868) and sensitivity/specificity to detect DR were highest for the indirect SBFI approach (0.79/0.99 for any DR and 1.0/1.0 for severe DR, 0.79/1.0 for diabetic maculopathy). CONCLUSIONS: Smartphone-based fundus imaging can meet DR screening requirements in an outreach setting; however, not all devices are suitable in terms of image quality and diagnostic accuracy. Smartphone-based fundus imaging might aid in alleviating the burden of DR screening in low- and middle-income countries, and these results will allow for a better selection of SBFI devices in field trials for DR screening.

Comparing low-cost handheld autorefractors: A practical approach to measuring refraction in low-resource settings.

PURPOSE: To compare and validate the accuracy and ease of use of handheld autorefractors against retinoscopic refraction by an ophthalmologist for assessing the visual acuity of older adults in India. METHODS: 190 patients were enrolled at the Sankara Eye Hospital in Bangalore, India, to undergo refraction using three different handheld devices-Retinomax (Nikon Inc., Japan), Netra (Eyenetra, Inc., USA), and QuickSee (PlenOptika, Inc., USA)-and the results were compared with cycloplegic retinoscopy and refraction done by an ophthalmologist. We analyzed the mean, standard deviation (S.D.), and Bland-Altman comparison of dioptric (D) power accuracy. RESULTS: The difference between the handheld devices and subjective refraction for each device was: Retinomax (N = 186), mean -0.41 D, S.D. 2.14; Netra (N = 179), mean 0.61 D, S.D. 2.20; and QuickSee (N = 182), mean -0.05 D, S.D. 1.04. CONCLUSION: The QuickSee and the Retinomax may be used successfully as refraction screening tools in epidemiologic studies of adults in India and as diagnostic tools in low-resource settings.

Spectacle Wear Among Children in a School-Based Program for Ready-Made vs Custom-Made Spectacles in India: A Randomized Clinical Trial.

Importance: Uncorrected refractive errors are the most common cause of visual impairment in children despite correction being highly cost-effective. Objective: To determine whether less expensive ready-made spectacles produce rates of spectacle wear at 3 to 4 months comparable to those of more expensive custom-made spectacles among eligible school-aged children. Design, Setting, and Participants: This noninferiority, double-masked, randomized clinical trial recruited children aged 11 to 15 years from January 12 through July 31, 2015, from government schools in urban and periurban areas surrounding Bangalore, India. Follow-up occurred from August 1 through September 31, 2015. Participants met the following eligibility criteria for ready-made spectacles: failed vision screening at the 6/9 level in each eye; refraction was indicated; acuity improved with correction by 2 or more lines in the better-seeing eye; the corrected acuity with the spherical equivalent was not more than 1 line less than with full correction; anisometropia measured less than 1.0 diopter; and an appropriate frame was available. Interventions: Eligible children were randomized to ready-made or custom-made spectacles. Main Outcomes and Measures: Proportion of children wearing their spectacles at unannounced visits 3 to 4 months after the intervention. Results: Of 23 345 children aged 11 to 15 years who underwent screening, 694 had visual acuity of less than 6/9 in both eyes, and 535 underwent assessment for eligibility. A total of 460 children (227 female [49.3%] and 233 male [50.7%]; mean [SD] age, 13.4 [1.3] years) were eligible for ready-made spectacles (2.0% undergoing screening and 86.0% undergoing assessment) and were randomized to ready-made (n = 232) or custom-made (n = 228) spectacles. Follow-up rates at 3 to 4 months were similar (184 [79.3%] in the ready-made group and 178 [78.1%] in the custom-made group). Rates of spectacle wear in the 2 arms were similar among 139 of 184 children (75.5%) in the ready-made arm and 131 of 178 children (73.6%) in the custom-made arm (risk difference, 1.8%; 95% CI, -7.1% to 10.8%). Conclusions and Relevance: Most children were eligible for ready-made spectacles, and the proportion wearing ready-made spectacles was not inferior to the proportion wearing custom-made spectacles at 3 to 4 months. These findings suggest that ready-made spectacles could substantially reduce costs for school-based eye health programs in India without compromising spectacle wear, at least in the short term. Trial Registration: isrctn.com Identifier: ISRCTN14715120.

Spectacle wearing in children randomised to ready-made or custom spectacles, and potential cost savings to programmes: study protocol for a randomised controlled trial.

BACKGROUND: Uncorrected refractive errors are the commonest cause of visual impairment in children, with myopia being the most frequent type. Myopia usually starts around 9 years of age and progresses throughout adolescence. Hyperopia usually affects younger children, and astigmatism affects all age groups. Many children have a combination of myopia and astigmatism. To correct refractive errors, the type and degree of refractive error are measured and appropriate corrective lenses prescribed and dispensed in the spectacle frame of choice. Custom spectacles (that is, with the correction specifically required for that individual) are required if astigmatism is present, and/or the refractive error differs between eyes. Spectacles without astigmatic correction and where the refractive error is the same in both eyes are straightforward to dispense. These are known as 'ready-made' spectacles. High-quality spectacles of this type can be produced in high volume at an extremely low cost. Although spectacle correction improves visual function, a high proportion of children do not wear their spectacles for a variety of reasons. The aim of this study is to compare spectacle wear at 3-4 months amongst school children aged 11 to 15 years who have significant, simple uncorrected refractive error randomised to ready-made or custom spectacles of equivalent quality, and to evaluate cost savings to programmes. The study will take place in urban and semi-urban government schools in Bangalore, India. The hypothesis is that similar proportions of children randomised to ready-made or custom spectacles will be wearing their spectacles at 3-4 months. METHODS/DESIGN: The trial is a randomised, non-inferiority, double masked clinical trial of children with simple uncorrected refractive errors. After screening, children will be randomised to ready-made or custom spectacles. Children will choose their preferred frame design. After 3-4 months the children will be followed up to assess spectacle wear. DISCUSSION: Ready-made spectacles have benefits for providers as well as parents and children, as a wide range of prescriptions and frame types can be taken to schools and dispensed immediately. In contrast, custom spectacles have to be individually made up in optical laboratories, and taken back to the school and given to the correct child. TRIAL REGISTRATION: ISRCTN14715120 (Controlled-Trials.com) Date registered: 04 February 2015.

Leber's congenital amaurosis as the retinal degenerative phenotype in thiamine responsive megaloblastic anemia: a case report.

BACKGROUND: Thiamine responsive megaloblastic anemia syndrome (TRMA), an autosomal recessive disorder is caused by mutations in the SLC19A2 gene which encodes for thiamine transporter 1 (THTR1) protein. TRMA presents with a triad of clinical features that includes diabetes mellitus, megaloblastic anemia and sensorineural hearing loss. Apart from the triad, reported ophthalmic features include cone rod dystrophy, optic atropy and retinitis pigmentosa. MATERIALS AND METHODS: A female child presented with Leber's congenital amaurosis at 10 months of age, later diagnosed with hearing impairment at 1 year, diabetes mellitus and megaloblastic anemia at 3 and a half years of age and hence as a case of thiamine responsive megaloblastic anemia. Six exons of the candidate gene SLC19A2 were screened by PCR and direct sequencing. SIFT and PolyPhen analysis was done to predict the probable effect of the mutation. RESULTS: Sequence analysis of the SLC19A2 coding region revealed a novel missense mutation in exon 2; c.314 G > A (p.G105E). Segregation analysis revealed parents heterozygous for the mutation and unaffected sib homozygous for wild type. SIFT and PolyPhen analyses predicted the mutation to be "damaging" (score-0.02) and "probably damaging" (score-0.994), respectively. CONCLUSIONS: SLC19A2, the high-affinity thiamine transporter, is the only gene known to be associated with TRMA. Here we describe for the first time Leber's congenital amaurosis as the retinal phenotype and also report a novel point mutation in the SLC19A2 gene that co-segregated with the disease in a TRMA patient.