Retinopathy of prematurity (ROP) remains one of the leading preventable causes of childhood blindness worldwide. In high-income countries, structured screening programs with widefield digital fundus imaging have dramatically reduced vision loss from this condition. However, in much of sub-Saharan Africa, South Asia, and Latin America, where a “third epidemic” of ROP is currently being experienced, driven by expanded neonatal intensive care without parallel investment in eye health infrastructure, the situation remains deeply unequal. A prospective cross-sectional study conducted at the Aravind Eye Hospital in Coimbatore, India, offers a compelling answer to a question that our specialty has long debated: Can affordable, smartphone-based imaging, combined with artificial intelligence, replace costly widefield camera systems that most low- and middle-income countries cannot afford? This study enrolled 156 premature infants (312 eyes; mean gestational age, 33 weeks) from January 2021 to April 2022. Each infant underwent fundus imaging using both widefield digital fundus imaging and one of two smartphone-based fundus imaging devices, the “Make In India Retcam” and the Keeler Monocular Indirect Ophthalmoscope. Two blinded graders then evaluated all images for zone, stage, plus disease, and vascular severity scores. Smartphone images were subsequently used to train a ResNet18 deep learning architecture to classify normal versus pre-plus/plus disease. The results were striking: human graders using smartphone-based images achieved 100% sensitivity and 83.49% specificity in detecting treatment-requiring ROP — performance that is clinically indistinguishable from premium widefield systems. The AI model matched the sensitivity (100%) with a specificity of 58.6%, pointing to the technology’s promise as a high-sensitivity triage tool capable of flagging every infant who truly needs intervention, even if it requires secondary human review for confirmation. No statistically significant difference was found between the two SBFI devices tested, reinforcing the generalizability of this approach across different affordable camera systems. WHAT THIS MEANS FOR OUR GLOBAL COMMUNITY AI as a bridge across the equity gap The significance of these findings extends beyond retinopathy of prematurity. They illustrate a broader principle that is already reshaping global health: when artificial intelligence is validated against robust clinical standards and deployed on accessible hardware, it can extend specialist-level diagnostic capacity to settings where pediatric ophthalmologists are scarce or entirely absent. Consider the arithmetic of inequality. In high-income countries, there is approximately one ophthalmologist for every 20,000 people. Across much of sub-Saharan Africa and Southeast Asia, that ratio often exceeds 1:500,000. No training program, however ambitious, will close this gap within the next decade. AI-assisted diagnostic tools will not replace pediatric ophthalmologists; however, they can make one expert’s knowledge available to thousands of infants born in district hospitals far from any specialist center. The Aravind study points to a tangible model in which a technician captures fundus images using an affordable smartphone adapter; an AI algorithm performs a preliminary severity assessment and flags cases requiring urgent review; and a remote specialist confirms the finding via a telemedicine platform. This three-step cascade can be implemented at a fraction of the cost of establishing a fully equipped screening unit and can reach rural and peri-urban populations that are currently outside any formal screening program. Beyond ROP: a platform for pediatric ocular surveillance The pipeline described above is disease-agnostic. The same technological architecture that screens for ROP can be adapted for amblyopia detection through red reflex analysis, for congenital glaucoma surveillance through automated optic disc assessment, for the identification of pediatric cataracts, and for the early recognition of leukocoria, the white pupillary reflex that may signal retinoblastoma, a diagnosis in which days can determine not only vision but also survival. In settings where population-level surveillance has never existed, these tools offer the prospect of genuine epidemiological mapping for the first time. Understanding the true burden of childhood blindness in regions that have historically been invisible to global datasets is not merely an academic exercise; it is a necessary precondition for rational resource allocation, advocacy, and policy change. The role of low- and middle-income countries in shaping the future of AI There is a tempting but dangerous narrative that positions resource-limited countries as passive recipients of technologies designed and validated in the Global North. The Aravind study actively challenges this framing. The research was conducted in India, led in part by Indian institutions, used a camera developed specifically for the Indian market, and was trained on a dataset drawn from Indian neonates. That matters. AI models are only as reliable as the populations on which they are trained, and a model built on images from North American or European neonatal units may perform very differently on infants of different ancestries, born at different gestational ages, and under different environmental conditions. The call to action for our community is clear: low- and middle-income countries must not only adopt these tools but also generate clinical data, participate in algorithm development, and contribute to validation studies. WSPOS, as a global society, is uniquely positioned to facilitate these collaborations by connecting centers of excellence in the Global South with research institutions, industry partners, and funding agencies capable of supporting this work. Challenges that demand honest acknowledgement The enthusiasm for AI in ophthalmology must be tempered by realism. Several challenges must be addressed before these technologies can achieve their full potential in resource-limited settings. Data sovereignty and algorithmic bias are genuine concerns. Algorithms trained predominantly on images from well-resourced centers may embed assumptions about image quality, camera type, or population characteristics, thereby reducing their accuracy when deployed elsewhere. Robust external validation across diverse populations is not optional; it is an ethical imperative. Infrastructure requirements are not negligible. Even the most affordable smartphone-based system requires reliable electricity, Internet connectivity for telemedicine, and trained personnel. In the most remote settings, these prerequisites may themselves be binding constraints. Regulatory frameworks in many low- and middle-income countries have not kept pace with the pace of AI development. Deploying diagnostic algorithms in the absence of clear guidance on liability, consent, and quality assurance creates risks for patients, practitioners, and institutions alike. Sustainability beyond pilot projects requires durable business models, government commitment, and integration into national screening programs, steps that demand political will and cross-sectoral partnerships. None of these challenges is insurmountable. However, acknowledging them honestly is the first step toward addressing them systematically. The World Society of Paediatric Ophthalmology and Strabismus recognizes artificial intelligence as one of the most significant opportunities in our specialty’s history, not because it will eliminate the need for skilled clinicians, but because it extends our reach to children who have never had access to pediatric ophthalmological care. We call on our member societies and individual colleagues to: • Engage actively with AI research, both as clinical collaborators and as authors of validation studies within their own patient populations. • Promote the incorporation of telemedicine and AI-assisted screening into national neonatal care protocols by actively advocating within health systems. • Encourage the creation of open-access datasets that encompass images from diverse populations to support the development of algorithms that are broadly applicable beyond their original contexts. • Support training initiatives that equip clinical staff in resource-limited settings with the skills to use AI-enabled screening tools both safely and effectively. • Participate in WSPOS-led working groups that develop international guidelines for the ethical deployment of AI in pediatric eye care. The technology is ready or nearly so. The question before our community is whether we have the collective will to ensure that it reaches every child who needs it.

Key learning points: This study evaluated health-related quality of life (HRQOL) in children with strabismus and examined: • How children’s self-reports compare with parent perceptions • Whether factors like age, sex, or type of strabismus influence outcomes after surgery A total of 114 children with strabismus and 29 controls were assessed using: • Eye-specific questionnaire (IXTQ) • General quality of life tool (PedsQL) and general questionnaire, which together constituted the HRQOL Main findings: • Children with strabismus had lower HRQOL than controls, even after surgery. • Surgery significantly improved HRQOL in eye-related, psychosocial, and physical domains. • Improvement was independent of age, sex, glasses use, or strabismus type. • There was overall good agreement between children and parents, but children rated their quality of life lower than their parents did. • Even post-surgery, HRQOL did not reach normal levels compared to healthy children. Why I selected the paper: This paper underlines the concept that strabismus affects more than vision. Strabismus has a significant psychosocial and quality-of-life impact on children. We tend to think of this aspect more in the adult population, but this is a reminder that it is an important consideration in the pediatric age group as well. Effects can persist even after surgical correction, so self-perceptions may remain low even after interventions, which may indicate a need for psychological or social support interventions. The benefits of surgery are broadly applicable, and improvement is not influenced by age, sex, type of strabismus, or use of glasses. Child and parent perception can differ, so the surgeon or physician should not rely solely on parent proxy reports. The child’s voice matters, and parent perception may not fully reflect the child’s experience.

Why this paper was selected: This paper was selected because it is the first paper which addresses a critical clinical question in pediatric ophthalmology: whether newer “enhanced” monofocal lenses—originally designed for adults—offer tangible benefits for children. It provides high-quality evidence through a prospective randomized study. The study is particularly relevant because: • Target population: It focuses specifically on children, a group for whom the standard of care has traditionally been monofocal lenses to ensure the best distance vision. • Technological comparison: It compares the standard lens to an enhanced version that uses a higher-order aspheric surface to improve intermediate vision without the common side effects of multifocal lenses, such as halos or glare. Key learning points: • Improved intermediate performance: The primary finding is that the enhanced monofocal (Eyhance) IOL significantly improves visual performance at intermediate distances (greater than 1 meter) compared to the standard lens. • Consistent distance vision: Both lens types provided comparable results for uncorrected and corrected distance visual acuity. The enhanced lens successfully improved depth of focus without “jeopardizing” distance vision. • Limitations in near vision: Despite the improvements at intermediate distances, the enhanced monofocal lens did not reduce the need for or the power of spectacles for near distances (less than 1 meter). • Superior defocus curve: When testing through a range of focus (simulating distances from 2 meters to 25 cm), children with the Eyhance lens had significantly better mean visual acuity (0.30 vs. 0.41 logMAR). This highlights the importance of using defocus curves rather than single-point intermediate testing to capture the true depth-of-focus benefits in children. • Quality of vision: The enhanced lens offers a superior depth of focus while avoiding the dysphotopsias, such as halos, typically associated with multifocal IOLs. The paper serves as a foundational study for future research, including: • Long-term visual outcomes: Further studies could investigate the long-term stability and visual performance of enhanced monofocal IOLs beyond the 3-month follow-up period used in this study. This would be particularly important to monitor for potential issues such as late-onset visual axis opacification in the pediatric population. • Expansion to younger age groups: This study focused on children aged 6 to 13 years. Future research could explore the safety and efficacy of these lenses in younger children, where visual development is more critical and the risk of amblyopia is higher. • Assessment of functional vision and quality of life: While this study measured objective visual acuity, future research could incorporate validated pediatric quality-of-life questionnaires to understand how improved intermediate vision impacts a child’s daily activities, such as schoolwork and digital device use. • Comparison with other IOL technologies: This study compared the Eyhance to a standard monofocal lens. Future research could provide direct head-to-head comparisons between enhanced monofocal IOLs and other options such as multifocal IOLs or non-diffractive extended depth of focus (EDOF) lenses in children.

The paper is clean, simple and helpful for practitioners.

I selected this systematic review because it was the research topic of my PhD. Neurofibromatosis type 1 (NF1) is a relatively common hereditary disease with a birth incidence of 1 in 2700 and an estimated prevalence of 1 in 4560 people worldwide. Patients are prone to develop both benign and malignant tumours of the central and peripheral nervous system. For timely diagnosis and treatment of these tumours, lifelong follow-up is required. Ophthalmologists play an important role in the surveillance of these patients, particularly in the early detection of optic pathway gliomas (OPGs). NF1-associated OPGs have a variable clinical course—some remain asymptomatic and require no intervention, while others may progress and lead to visual or systemic complications. There remains ongoing debate between centres regarding whether routine MRI screening should be performed in asymptomatic children. Key learning points: • 38 studies met the inclusion criteria. • The pooled prevalence of OPG in children and adolescents under 18 years of age with NF1 was 17% (95% CI, 14%–20%) using a random-effects model. • No statistically significant association was found between MRI surveillance strategies and OPG detection, reinforcing that there is no clear detection advantage with routine MRI screening. This study provides the first pooled estimate of OPG prevalence in children and adolescents with NF1 and supports the value of a symptom-based MRI surveillance approach.

This paper assessed dynamic depth perception (stereopsis) in adults with treated anisometropic amblyopia. It demonstrates that even after visual acuity is restored, binocular deficits can persist well into adulthood. These findings highlight the need to address binocular impairments during treatment in childhood. Improving binocularity in amblyopia may help alleviate ocular motor and visuomotor deficits associated with impaired binocular vision. Key learning points: 1. Restoring visual acuity does not restore binocular vision. 2. Dynamic stereoacuity deficits persist independently of anisometropia or static stereoacuity. 3. Amblyopia treatment must evolve beyond visual acuity restoration alone.

This paper was selected for its scientific rigor in comparing two pediatric anesthesia approaches and for presenting an alternative that can be readily incorporated into clinical practice. The study addresses important considerations in pediatric anesthesia, including discomfort during propofol administration, the need for improved postoperative analgesia, and achieving a shorter recovery period. Esketamine also has anti-inflammatory properties and may reduce the incidence of emergence agitation. Study design: Randomized controlled trial of pediatric patients undergoing strabismus surgery. Comparison: • Propofol alone • Low-dose esketamine combined with propofol Key findings: 1. Improved hemodynamic stability with fewer intraoperative hypotension episodes. 2. Reduced total propofol requirements (propofol-sparing effect). 3. Lower postoperative pain scores and reduced need for additional analgesics. 4. Smoother recovery with reduced emergence agitation. 5. No significant increase in respiratory complications or severe adverse events. Overall, the combination offers improved anesthesia quality without compromising safety.

Myopia is on the rise, and with predictions suggesting that its prevalence may double by 2050, it has become a global health concern. As myopia is multifactorial, extensive research is ongoing to study the contributory factors associated with its development and progression. One study that caught my attention was a paper titled, “A Novel Comparative Study of Inflammatory Cytokines Through Noninvasive Tear Analysis in Children with Myopia Versus Emmetropia” by Nishanth S. et al., published in the American Journal of Ophthalmology (2025). In this study, the authors hypothesized the concept of “para-inflammation,” a subthreshold inflammatory response, as a possible contributor to childhood myopia. They evaluated the OSDI (Ocular Surface Disease Index) questionnaire, tear meniscus height (TMH), tear film break-up time (TBUT), and inflammatory cytokines (IL-1β, IL-6, IL-10, MMP-9, ICAM-1, IL-17A, TNF-α, and VEGF-A) in the tear fluid of myopic children and compared them with hyperopic and emmetropic children. The study demonstrated a positive trend in all inflammatory cytokines except IL-17 and VEGF-A, with significantly increased levels of TNF-α and ICAM-1 in the tear film of myopes. Both molecules are classical mediators of ocular surface inflammation, and their elevated levels in tears suggest a low-grade inflammatory response and a probable link to myopia. In addition, children with moderate and high myopia showed lower TMH and TBUT values, and those with moderate myopia had significantly higher OSDI scores. In summary, this study highlights an interplay between dry eye, low-grade inflammation, and myopia. The emerging evidence from this study suggests the need to move beyond purely optical and environmental factors. This expands the conventional approach—from simply correcting refractive error and prescribing low dose atropine eyedrops or defocus glasses—to adopting a more holistic strategy that includes assessment of ocular surface health, lifestyle evaluation, and optimization of inflammatory status. In daily clinical practice, attention should also be directed toward evaluating tear film stability and dry eye parameters in myopic children, including routine assessment of TBUT, meibomian gland function, and blink patterns (particularly in high screen users), along with age-appropriate OSDI questionnaires to enhance comprehensive myopia care. Ocular surface optimization—through lubricants, management of meibomian gland dysfunction (MGD), blink training, and inflammation control when indicated—may be particularly relevant before initiating therapies such as orthokeratology or contact lens wear. Additionally, the role of the lid margin microbiome in ocular inflammation and pediatric MGD warrants further exploration. Although tear cytokine testing is not yet clinically applicable due to cost and lack of standardization, the concept that dry eye signs and inflammation may influence axial elongation has practical counseling value. Appropriate blinking and structured lifestyle advice—limiting recreational screen time (<2 hours/day), encouraging ≥2 hours/day of outdoor activity, maintaining an appropriate reading distance (30–40 cm), and following the 20-20-20 rule—should form an integral part of myopia management. Furthermore, recognition that atropine may exert anti-inflammatory effects, in addition to its established role in myopia control, provides additional reassurance in its use for progressive myopia. While this study does not yet mandate changes in therapeutic protocols, it reinforces the importance of integrating ocular surface evaluation and lifestyle modification with blinking into routine pediatric myopia management. Ultimately, this evidence supports a multifactorial, preventive, and personalized approach to myopia care rather than a purely refractive model, and it highlights the potential of tear analysis as a non-invasive method for identifying biomarkers.