Building and maintaining interprofessional collaborative practice in eyecare: Learnings from the Queensland Paediatric Optometry Alignment Program.

PURPOSE: The Queensland Children's Hospital Paediatric Optometry Alignment Program commenced with a pilot phase to assess its feasibility, effectiveness and acceptability. This study identified the barriers that hinder effective interprofessional collaboration and the facilitators that contribute to its success, and assessed changes in optometrists' satisfaction since the pilot phase of the collaborative care programme. METHODS: Qualitative deductive and inductive content analysis was applied to open-ended free-text survey responses collected in 2018 from the optometrists involved in the Program's pilot phase. The responses were coded using the Theoretical Domains Framework (TDF) to categorise barriers and facilitators into key themes. Key behavioural determinants were mapped to the COM-B (Capability, Opportunity, Motivation-Behaviour) elements of the Behaviour Change Wheel model to identify intervention strategies. Intervention recommendations were derived from behaviour change mapping and compared with programme quality improvement initiatives. A cross-sectional explanatory survey informed by the TDF was conducted within the current 2023 cohort, and a longitudinal comparative analysis was carried out using data from the 2018 survey. RESULTS: Among the 97 surveys distributed in 2018, 44 respondents participated; from this group, 38 individuals contributed a total of 200 free-text responses. Facilitators (240 comments) outnumbered barriers (65 comments). Key facilitators were accessible and timely care, professional development, confidence and positive outcome beliefs. Barriers included communication, information handover, credibility, relationships and skill gaps. Optometrists actively engaged in the programme in 2023 reported heightened satisfaction with their involvement, increased confidence and greater engagement in paediatric eyecare delivery. However, challenges in clinical information transfer persist. CONCLUSION: The interprofessional collaborative model of paediatric eyecare has contributed efficiencies within the health system by building paediatric care capacity in the community, fostering professional credibility and promoting interdisciplinary trust. Insights gained should prove valuable for other paediatric eyecare services exploring hospital-to-community care models.

Deep Learning Approach for Differentiating Etiologies of Pediatric Retinal Hemorrhages: A Multicenter Study.

Retinal hemorrhages in pediatric patients can be a diagnostic challenge for ophthalmologists. These hemorrhages can occur due to various underlying etiologies, including abusive head trauma, accidental trauma, and medical conditions. Accurate identification of the etiology is crucial for appropriate management and legal considerations. In recent years, deep learning techniques have shown promise in assisting healthcare professionals in making more accurate and timely diagnosis of a variety of disorders. We explore the potential of deep learning approaches for differentiating etiologies of pediatric retinal hemorrhages. Our study, which spanned multiple centers, analyzed 898 images, resulting in a final dataset of 597 retinal hemorrhage fundus photos categorized into medical (49.9%) and trauma (50.1%) etiologies. Deep learning models, specifically those based on ResNet and transformer architectures, were applied; FastViT-SA12, a hybrid transformer model, achieved the highest accuracy (90.55%) and area under the receiver operating characteristic curve (AUC) of 90.55%, while ResNet18 secured the highest sensitivity value (96.77%) on an independent test dataset. The study highlighted areas for optimization in artificial intelligence (AI) models specifically for pediatric retinal hemorrhages. While AI proves valuable in diagnosing these hemorrhages, the expertise of medical professionals remains irreplaceable. Collaborative efforts between AI specialists and pediatric ophthalmologists are crucial to fully harness AI's potential in diagnosing etiologies of pediatric retinal hemorrhages.

Development of a joint set of database parameters for the EU-ROP and Fight Childhood Blindness! ROP Registries.

BACKGROUND/AIMS: The incidence of retinopathy of prematurity (ROP) is increasing and treatment options are expanding, often without accompanying safety data. We aimed to define a minimal, patient-centred data set that is feasible to collect in clinical practice and can be used collaboratively to track and compare outcomes of ROP treatment with a view to improving patient outcomes. METHODS: A multinational group of clinicians and a patient representative with expertise in ROP and registry development collaborated to develop a data set that focused on real-world parameters and outcomes that were patient centred, minimal and feasible to collect in routine clinical practice. RESULTS: For babies receiving ROP treatment, we recommend patient demographics, systemic comorbidities, ROP status, treatment details, ophthalmic and systemic complications of treatment, ophthalmic and neurodevelopmental outcomes at initial treatment, any episodes of retreatment and follow-up examinations in the short and long-term to be collected for use in ROP studies, registries and routine clinical practice. CONCLUSIONS: We recommend these parameters to be used in registries and future studies of ROP treatment, to reduce the variation seen in previous reports and allow meaningful assessments and comparisons. They form the basis of the EU-ROP and the Fight Childhood Blindness! ROP Registries.

Multicenter Validation of Deep Learning Algorithm ROP.AI for the Automated Diagnosis of Plus Disease in ROP.

Purpose: Retinopathy of prematurity (ROP) is a sight-threatening vasoproliferative retinal disease affecting premature infants. The detection of plus disease, a severe form of ROP requiring treatment, remains challenging owing to subjectivity, frequency, and time intensity of retinal examinations. Recent artificial intelligence (AI) algorithms developed to detect plus disease aims to alleviate these challenges; however, they have not been tested against a diverse neonatal population. Our study aims to validate ROP.AI, an AI algorithm developed from a single cohort, against a multicenter Australian cohort to determine its performance in detecting plus disease. Methods: Retinal images captured during routine ROP screening from May 2021 to February 2022 across five major tertiary centers throughout Australia were collected and uploaded to ROP.AI. AI diagnostic output was compared with one of five ROP experts. Sensitivity, specificity, negative predictive value, and area under the receiver operator curve were determined. Results: We collected 8052 images. The area under the receiver operator curve for the diagnosis of plus disease was 0.75. ROP.AI achieved 84% sensitivity, 43% specificity, and 96% negative predictive value for the detection of plus disease after operating point optimization. Conclusions: ROP.AI was able to detect plus disease in an external, multicenter cohort despite being trained from a single center. Algorithm performance was demonstrated without preprocessing or augmentation, simulating real-world clinical applicability. Further training may improve generalizability for clinical implementation. Translational Relevance: These results demonstrate ROP.AI's potential as a screening tool for the detection of plus disease in future clinical practice and provides a solution to overcome current diagnostic challenges.

Incomplete peripheral retinal vascularisation in retinopathy of prematurity: is it the consequence of changing oxygen saturation?

Background: We wish to determine the prevalence and risk factors of incomplete peripheral avascular retina (IPAR) in children screened for retinopathy of prematurity (ROP) and its association with oxygen saturation (SpO2) targets. Methods: A retrospective review of retinal images of premature infants born and screened for ROP in Auckland Region, New Zealand, between January 2013 and December 2017 was conducted. Images were reviewed to determine if avascular retina was present at their final ROP screening. The prevalence of peripheral avascular retina was compared among infants born prior to (Group 1) and after (Group 2) 2015 when the SpO2 target was increased. Infants with any concurrent ocular pathology or who had received ROP treatment were excluded. Results: In total, 62 (12.8%) of the total of 486 infants (247 in Group 1; 239 in Group 2) were found to have IPAR at their last ROP screening. Group 1 had more statistically significant infants with IPAR compared to Group 2 (39/247 infants and 23/239 infants respectively; p = 0.043). Conclusions: Incomplete peripheral retinal vascularisation occurred at a prevalence of 12.8% in infants at risk of ROP. Higher SpO2 targets did not increase the prevalence of incomplete peripheral retinal vascularisation. Low gestational age and low birth weight are likely risk factors for the development of avascular retina. Further research into the risk factors associated with incomplete peripheral retinal vascularisation and the associated long-term outcomes is needed.

Increased risk of retinopathy of prematurity since increased O2 saturation targets: A multi-centre study.

BACKGROUND/AIMS: Retinopathy of prematurity (ROP) is a leading cause of visual impairment in premature neonates. The BOOST II, SUPPORT and COT trials recommended increasing O2 saturation targets for pre-term neonates to reduce mortality; however, this is a risk factor for ROP. We aimed to determine whether these targets increased prevalence of ROP among pre-term neonates and higher risk groups. METHODS: Retrospective cohort study conducted using data from the Australian and New Zealand Neonatal Network. 17 298 neonate cohort born 2012-2018 at <32 weeks' GA and/or <1500 g BW was analysed. Adjusted odds ratios (aORs) were calculated for post-2015 risk of: any ROP; ROP ≥ Stage 2; and treated ROP. Sub-analysis stratified at <28 GA, < 26 weeks' GA, <1500 g BW and <1000 g BW was performed. RESULTS: Risk of any ROP increased in the post-2015 group (aOR = 1.23, 95% confidence interval (CI) = 1.14-1.32), <28 weeks' GA (aOR = 1.31, 95% CI = 1.17-1.46), <26 weeks (aOR = 1.57, 95% CI = 1.28-1.91), <1500 g (aOR = 1.24, 95% CI = 1.14-1.34) and <1000 g (aOR = 1.34, 95% CI = 1.20-1.50). ROP ≥ Stage 2 increased at <28 weeks (aOR = 1.30, 95% CI = 1.16-1.46), <26 weeks (aOR = 1.57, 95% CI = 1.28-1.91), <1500 g (aOR = 1.18, 95% CI = 1.08-1.30), and <1000 g (aOR = 1.26, 95% CI = 1.13-1.42). CONCLUSION: O2 therapy guidelines since 2015 have resulted in decreased mortality but increased risk of ROP. Individualised NICU adjustments of ROP screening/follow-up methods are necessary to address the clinical burden.

Clinical approach for suspected optic disc swelling in children: recommendations based on a six-year review.

CLINICAL RELEVANCE: Suspected optic disc swelling is a common presentation in children. The delineation between true optic disc swelling and pseudopapilloedema, its common masquerade, requires careful evaluation. A streamlined pathway is required to avoid unnecessary investigations. BACKGROUND: Papilloedema requires urgent neuroimaging, however, perceived optic disc swelling is not always true papilloedema. This study aims to investigate the outcome of referrals for optic disc swelling and formulate features that may assist in investigation and diagnosis. METHODS: A retrospective review of referrals for optic disc swelling to the Queensland Children's Hospital, Australia, between January 2014 and June 2020 was undertaken. RESULTS: Four hundred and ten children were referred for optic disc swelling. Sixty-six patients were confirmed with optic disc swelling, and 344 patients had pseudopapilloedema. The average age was 10.10 ± 3.57 and 9.90 ± 3.50 years, respectively. The most common aetiology of optic disc swelling was idiopathic intracranial hypertension (n = 25). Optic disc drusen constituted the majority of pseudopapilloedema (n = 239) and the remainder were crowded/tilted discs (n = 105). True optic disc swelling patients were more likely to experience headache (OR = 8.68, p < 0.01) and visual disturbance (OR = 2.14, p = 0.03). B-scan was the most sensitive for the detection of optic disc drusen (100%), followed by optical coherence tomography (70.38%) and fundus autofluorescence (44.86%). The retinal nerve fibre layer thickness was significantly thicker in true optic disc swelling compared to pseudopapilloedema (p < 0.01). Twenty-two (33.33%) true optic disc swelling patients and 33 (9.59%) pseudopapilloedema patients underwent neuroimaging prior to ophthalmology review. CONCLUSIONS: Suspected optic disc swelling in children is most likely pseudopapilloedema. Referrals should include neurological assessment, visual acuity, fundus photography, and optical coherence tomography to assist in the triage for ophthalmic review. Neuroimaging prior to ophthalmic review should be discouraged for children without neurological symptoms.

Positive predictive value of dual-modality vision screening in school children 4-7 years of age-a retrospective review in Queensland, Australia.

PURPOSE: To report the outcomes and positive predictive value (PPV) of vision screening in schoolchildren 4-7 years of age through the Primary School Nurse Health Readiness Program (PSNHRP) in Queensland, Australia. METHODS: A retrospective review of schoolchildren who underwent vision screening between January 2017 and December 2020 was conducted. Vision screening was performed through a dual-examination method, using the Parr 4m letter-matching vision test with crowding bars and the Spot photoscreener. Children were referred to an optometrist or ophthalmologist for review as required and if they failed either screening modality. PPVs were calculated based on whether a visual abnormality was confirmed by an ophthalmologist or optometrist. RESULTS: Of 185,685 eligible children, 176,164 (94.9%) consented to vision screening, 164,890 (93.6%) consented children underwent vision screening, and of those 12,148 (7.4%) were referred for an eye assessment. Of the 8,659 children with a known outcome (71.3% of referred), 6,011 (69.4% of known outcomes) had a confirmed visual abnormality and 2,648 (30.6%) children did not. The PPV was 0.73 when a referral was indicated by the photoscreener result, 0.76 when indicated by visual acuity testing, and 0.91 when indicated by both the photoscreener and visual acuity testing. CONCLUSIONS: The PSNHRP vision screening program showed a high uptake, and the dual screening method was effective in identifying visual abnormalities, with higher PPV when both visual acuity and photoscreener results indicated a need for referral.

Comparison of biometric and refractive changes in intermittent exotropia with and without overminus lens therapy.

BACKGROUND: Overminus lens is an effective non-surgical treatment of intermittent exotropia (IXT). Whether this treatment causes development of myopia is recently debated. This study evaluated ocular biometric changes in IXT children treated with and without overminus lens. METHODS: Children with IXT were randomly assigned to control and overminus groups. Cycloplegic refraction, axial length and other parameters (keratometry, anterior chamber depth and lens thickness) were evaluated for up to 30 months and compared between the groups. Axial length to corneal curvature (AL/CR) ratio was examined as an indicator of myopia progression. Changes prior to and after overminus were also assessed in a subgroup. RESULTS: Mean age of the total 84 subjects (58% women) was 7.2 years (range 4-15 years). Baseline refractive and biometric parameters of 39 control and 45 overminus subjects were not significantly different. Mean±SD change in refraction was -0.34±0.45D in controls and -0.41±0.66D in overminus group (p=0.527). AL increased by 0.29±0.20 millimetre (mm) in controls and 0.28±0.23 mm in overminus group (p=0.766). In the subgroup, the changes before and after overminus therapy was not significantly different (p>0.05). AL/CR ratio was maintained from baseline to follow-up visit (p=0.298) in both groups. CONCLUSION: No significant differences in biometry or refraction were found between the two groups and before and after the therapy. We conclude that the overminus therapy in children, not accounting for other myopia risk factors, does not affect normal refractive growths.

The paediatric optometry alignment program - a model of interprofessional collaborative eyecare.

CLINICAL RELEVANCE: Collaboration  between hospital-based ophthalmology and community-based optometry could pave the way to improve access to paediatric eyecare services. BACKGROUND: The Paediatric Optometry Alignment Program (POAP) began in 2016 as a proof-of-concept pilot project that aimed to improve access to specialist paediatric ophthalmology services. If found to be effective at improving patient access, and the quality of care acceptable to patients and professionals then the strategic intent was to upscale the programme to  serve as a model for paediatric eye care in the community. METHODS: Temporal observational trend analysis was used to review ophthalmology clinic appointment waitlists prior and post POAP pilot project commencement. Family satisfaction with post-discharge care was surveyed in a purposive sample of 30 patients.  Aligned optometrists in the program pilot (n = 97) were invited to complete an online survey (response rate 46%; n = 45). RESULTS: The percentage of children waiting longer than clinically recommended fell from 72% to 36%. Sixty-seven percent  of surveyed families had attended a community optometrist as recommended, and all rated the optometry experience from good to excellent.  Participating optometrists reported high levels of satisfaction with involvement in the program, and increased confidence and involvement in paediatric eye care delivery. The need to improve formal transfer  of clinical information was identified. CONCLUSION: Facilitated integrated care between community-based optometrists and a hospital-based ophthalmology department can improve access for tertiary care services, with high satisfaction for families and participating community-based optometrists.

Ocular manifestations of systemic diseases in children.

Knowledge of ocular diseases and understanding of the complex interplay between eye and systemic health have increased over the years. This knowledge is particularly important when caring for our youngest and most vulnerable paediatric patients when ophthalmic manifestations may provide an insight to underlying systemic diseases and can act as the first indicator of an undiagnosed systemic condition. Further, the visual system can be vulnerable to manifestations of known systemic disease, with vigilant ophthalmic examination generally aiding early identification of ocular complications for collaborative multidisciplinary care to prevent avoidable vision loss. The potential ocular signs and complications of the following developmental, genetic or acquired childhood systemic disorders are presented: premature birth, trisomy 21, albinism, Marfan's syndrome, Stickler's syndrome, septo-optic dysplasia, aniridia, neurofibromatosis 1, Sturge-Weber syndrome, papilloedema, juvenile idiopathic arthritis and vitamin A deficiency. Rather than providing an exhaustive list of diseases, this review offers an overview of the more commonly encountered congenital or acquired childhood systemic conditions that have associated childhood ophthalmic disorders and presents referral and ongoing surveillance recommendations.

Validity and repeatability of contour-based visotec distance stereoacuity test.

CLINICAL RELEVANCE: The clinical assessment of distance stereoacuity is important in some ocular conditions. Given the different neurophysiological mechanism for crossed and uncrossed stereoacuities, evaluation of both may provide additional insight into binocular vision disorders. BACKGROUND: Clinical devices measuring distance crossed and uncrossed stereopsis are not readily available. Visotec Distance Stereo Test (VDST) is a contour-based device designed to measure both forms of distance stereoacuity. This study assesses the validity and test-retest reliability of the device in comparison to the random dot-based Randot Distance Stereo Test (RDST). METHOD: VDST and RDST were administered to a total of 107 children, that included 51 'normal' and 56 'abnormal' (37 intermittent exotropia and 19 amblyopia) between the ages of 5 and 15 years. Two examiners retested stereoacuities in a sub-set of 62 randomly selected subjects. Stereoacuity was transformed to log scale. 95% limits of agreement were calculated for test-retest reliability. The Bland-Altman plot was used to demonstrate the agreement between the tests and the examiners. RESULTS: The mean ± SD crossed distance stereoacuities using VDST in normal, intermittent exotropia and amblyopic children were 93.1 ± 43.8, 161.9 ± 89.8 and 236.3 ± 122.4 arcseconds, respectively. For uncrossed stereoacuity, these were 104.7 ± 54.0, 187.6 ± 89.6 and 265.5 ± 144.0 arcseconds, respectively. Crossed stereoacuity was significantly better than uncrossed stereoacuity. 95% limits of test-retest agreement for crossed and uncrossed stereoacuities using the VDST were 0.27 and 0.30, respectively. An exact test-retest match using VDST was 84% in normals and 77% in abnormals for crossed and 83% in normal and 74% in abnormal for uncrossed stereoacuities. CONCLUSION: VDST is a reliable, valid and easy-to-administer distance crossed and uncrossed stereoacuity measuring device. Further studies are required to establish the clinical importance of assessing these two forms stereoacuities in relation to various binocular vision disorders.

Long-term visual outcomes of children screened for retinopathy of prematurity with telemedicine in New Zealand.

CLINICAL RELEVANCE: Children with a history of regressed retinopathy of prematurity (ROP) are at increased risk of peripheral avascular retina. Wide-field digital retinal imaging and telemedicine is an effective tool for ROP screening. Ophthalmologists and Optometrists should have a high level of clinical suspicion for peripheral retinal changes in children screened for ROP. BACKGROUND: Retinopathy of prematurity, a vaso-proliferative disorder of the pre-term retina, is a preventable cause of childhood visual impairment. The Auckland Regional Telemedicine ROP (ART-ROP) network, established in 2006, utilises wide-field digital imaging and telemedicine to screen at-risk infants for ROP. This prospective observational study reports the long-term ocular outcomes of ART-ROP network infants. METHODS: A comprehensive paediatric eye examination including cycloplegic autorefraction and wide-field retinal imaging was completed on all participants. Participants had been screened for ROP by the ART-ROP network between May 2008 and October 2011. RESULTS: A total of 69 children, with a mean age of 5 to 8 years old were assessed and divided into two groups: those with or without a history of ROP, 44 and 25 children, respectively. Infants with a history of ROP had significantly lower gestational age (26.6 ± 1.9 vs. 29.1 ± 1.6 weeks, p < 0.001) and birth weight (937 ± 237 vs. 1177 ± 311 grams, p = 0.001). No significant differences were detected between the two groups for visual acuity (p = 0.596), stereopsis (p = 0.219), refractive error (p = 0.472), or strabismus. Clinically significant refractive error was noted in 10 participants; none with moderate or high myopia. Retinal imaging exposed asymptomatic, persistent, peripheral avascular retina in four children, all of whom had a history of regressed ROP. CONCLUSION: Visual and ocular outcomes did not vary based on history of ROP, with no participant having reduced vision as a result of undetected or untreated ROP. Further research is required into the long-term implication of persistent avascular retina in regressed ROP.

Assessment of distance-near control disparity in basic and divergence excess paediatric intermittent exotropia.

CLINICAL RELEVANCE: In intermittent exotropia (IXT), deviation is better controlled at near fixation. Understanding of the mechanism responsible for this common observation may improve clinical management of IXT. BACKGROUND: The physiological basis for the distance-near difference in control of deviation in IXT is vastly undetermined. A new parameter, 'control score disparity (CSD)', defined as the difference between distance and near control scores, is introduced. Association of CSD with positive fusional amplitude (PFA), accommodative convergence to accommodation (AC/A) ratio and distance angle of deviation was investigated to further understand the mechanisms. METHOD: Patients aged between four and fifteen years with basic and divergence excess IXT were included. Subjects with previous strabismus surgery, amblyopia and inability to perform clinical tests were excluded. A standardized office-based scoring system was used to assess IXT controls. Subjects were sub-divided into group 1 (CSD <2) and group 2 (CSD ≥2). Pearson's univariate and regression analysis were used to determine relationships between CSD and other independent variables. RESULTS: Mean age of the total 141 subjects (57.6% female) was 6.8 ± 2.5 years. Basic IXT was more common (60%) and 60% had CSD ≥2. The mean±SD distance angle of deviation, AC/A ratio, PFA and CSD were 22.1 ± 6.6 prism dioptres, 5.0 ± 1.0, 28.6 ± 6.3 prism dioptre and 2.0 ± 0.5, respectively. CSD was significantly correlated to PFA (r = 0.64, p < 0.001) and AC/A ratio (r = 0.27, p < 0.001) in overall samples and Group 2 subjects (r = 0.41, p = 0.001). CONCLUSIONS: PFA is a major factor associated with the distance/near difference of IXT control. Individuals with a higher PFA demonstrated greater CSD. While AC/A ratio was associated in better control at near in divergence excess IXT, magnitude of angle appears irrelevant.

Performance of deep-learning artificial intelligence algorithms in detecting retinopathy of prematurity: A systematic review.

PURPOSE: Artificial intelligence (AI) offers considerable promise for retinopathy of prematurity (ROP) screening and diagnosis. The development of deep-learning algorithms to detect the presence of disease may contribute to sufficient screening, early detection, and timely treatment for this preventable blinding disease. This review aimed to systematically examine the literature in AI algorithms in detecting ROP. Specifically, we focused on the performance of deep-learning algorithms through sensitivity, specificity, and area under the receiver operating curve (AUROC) for both the detection and grade of ROP. METHODS: We searched Medline OVID, PubMed, Web of Science, and Embase for studies published from January 1, 2012, to September 20, 2021. Studies evaluating the diagnostic performance of deep-learning models based on retinal fundus images with expert ophthalmologists' judgment as reference standard were included. Studies which did not investigate the presence or absence of disease were excluded. Risk of bias was assessed using the QUADAS-2 tool. RESULTS: Twelve studies out of the 175 studies identified were included. Five studies measured the performance of detecting the presence of ROP and seven studies determined the presence of plus disease. The average AUROC out of 11 studies was 0.98. The average sensitivity and specificity for detecting ROP was 95.72% and 98.15%, respectively, and for detecting plus disease was 91.13% and 95.92%, respectively. CONCLUSION: The diagnostic performance of deep-learning algorithms in published studies was high. Few studies presented externally validated results or compared performance to expert human graders. Large scale prospective validation alongside robust study design could improve future studies.

Retinopathy of prematurity screening: A narrative review of current programs, teleophthalmology, and diagnostic support systems.

PURPOSE: Neonatal care in middle-income countries has improved over the last decade, leading to a "third epidemic" of retinopathy of prematurity (ROP). Without concomitant improvements in ROP screening infrastructure, reduction of ROP-associated visual loss remains a challenge worldwide. The emergence of teleophthalmology screening programs and artificial intelligence (AI) technologies represents promising methods to address this growing unmet demand in ROP screening. An improved understanding of current ROP screening programs may inform the adoption of these novel technologies in ROP care. METHODS: A critical narrative review of the literature was carried out. Publications that were representative of established or emerging ROP screening programs in high-, middle-, and low-income countries were selected for review. Screening programs were reviewed for inclusion criteria, screening frequency and duration, modality, and published sensitivity and specificity. RESULTS: Screening inclusion criteria, including age and birth weight cutoffs, showed significant heterogeneity globally. Countries of similar income tend to have similar criteria. Three primary screening modalities including binocular indirect ophthalmoscopy (BIO), wide-field digital retinal imaging (WFDRI), and teleophthalmology were identified and reviewed. BIO has documented limitations in reduced interoperator agreement, scalability, and geographical access barriers, which are mitigated in part by WFDRI. Teleophthalmology screening may address limitations in ROP screening workforce distribution and training. Opportunities for AI technologies were identified in the context of these limitations, including interoperator reliability and possibilities for point-of-care diagnosis. CONCLUSION: Limitations in the current ROP screening include scalability, geographical access, and high screening burden with low treatment yield. These may be addressable through increased adoption of teleophthalmology and AI technologies. As the global incidence of ROP continues to increase, implementation of these novel modalities requires greater consideration.