Related Experiment Video
Updated: Aug 14, 2026
Using an Automated Hirschberg Test App to Evaluate Ocular Alignment
Published on: March 24, 2020
Clinical Performance and Implementation of AI-Enabled Paediatric Ophthalmic Screening, Triage, Diagnosis, and
Joel Somerville1, Mohammad Hussein Mustafa2, Mohamed Mahmoud Seweid3
1Centre for Rural Health Science, University of the Highlands and Islands, Inverness IV2 3JH, UK.
Insights
Artificial intelligence (AI) shows promise for pediatric eye care screening and triage, but further validation is needed. Current AI tools require human oversight and are not yet ready for autonomous implementation in routine practice.
Area of Science:
- Ophthalmology
- Artificial Intelligence
- Pediatric Healthcare
Background:
- AI is increasingly used in adult ophthalmology, but its role in pediatric eye care is less established.
- This review synthesizes evidence on AI tools for pediatric ophthalmic diagnosis, screening, triage, surveillance, and referral.
Purpose of the Study:
- To evaluate the diagnostic performance, safety, workflow integration, equity, and implementation readiness of AI tools in pediatric eye care.
- Emphasis on primary, community, and primary care-relevant settings.
Main Methods:
- PRISMA-guided systematic review of studies on AI/machine-learning tools for children (0-18 years) with eye conditions.
- Searched MEDLINE, Embase, Web of Science, Scopus, IEEE Xplore (inception to March 30, 2026).
- Assessed study quality using adapted frameworks (APPRAISE-AI, QUADAS-2, PROBAST+AI, ROBINS-I).
Main Results:
- Twelve studies included, covering retinopathy of prematurity, retinoblastoma, amblyopia, myopia, cataract, and visual acuity.
- AI demonstrated promising performance in diagnosis/screening (e.g., retinopathy of prematurity, retinoblastoma, myopia).
- Feasibility shown in various workflows (neonatal, school, community) using smartphone imaging and task-shifted operators, but validation and fairness assessments were limited.
Conclusions:
- AI tools show potential for pediatric ophthalmic screening, triage, and surveillance with human oversight.
- Confidence in AI performance is limited by study design, sample limitations, and lack of external/temporal validation.
- Routine autonomous implementation of AI in pediatric eye care is premature.
Abstract:
Background/Objectives: Artificial intelligence (AI) is increasingly being evaluated for ophthalmic diagnosis, screening, and triage, yet its role in paediatric eye care remains less established than in adult ophthalmology. This systematic review aimed to synthesise evidence on AI-enabled tools for paediatric ophthalmic diagnosis, screening, triage, surveillance, and referral, with an emphasis on diagnostic performance, safety, workflow integration, equity, and implementation readiness in primary, community, and primary care-relevant settings. Methods: A PRISMA-guided systematic review was conducted using MEDLINE, Embase, Web of Science, Scopus, and IEEE Xplore from inception to 30 March 2026. Eligible studies evaluated AI or machine-learning tools for children and adolescents aged 0-18 years in relation to paediatric eye conditions. Study selection and data extraction were undertaken independently by reviewers, with disagreements resolved by consensus or third-reviewer adjudication. Methodological and reporting quality was evaluated using an author-adapted six-domain rubric informed by APPRAISE-AI. Diagnostic-accuracy studies were assessed using an author-adapted QUADAS-2 framework incorporating QUADAS-AI-informed AI-specific considerations, the prediction-model study was assessed using PROBAST+AI, and the non-randomised treatment-effect study was assessed using ROBINS-I. The public dataset descriptor was evaluated separately using an author-developed dataset-quality, representativeness, and applicability framework. Because of clinical and methodological heterogeneity, findings were synthesised thematically. Results: Twelve empirical studies and one public dataset descriptor were included, covering retinopathy of prematurity, retinoblastoma, amblyopia risk, myopia, congenital cataract, and visual-acuity assessment. AI systems frequently demonstrated promising diagnostic or screening performance, including sensitivity-first detection of treatment-requiring retinopathy of prematurity, high discrimination for retinoblastoma activity, and strong myopia prediction using fundus images. Several studies supported feasibility in neonatal, school, and community workflows using smartphone-based imaging, task-shifted operators, tele-referral, and human-in-the-loop review. However, external and temporal validation, calibration, patient-level reporting, subgroup and fairness assessment, and economic evaluation were limited. Conclusions: AI-enabled tools show promise for supporting selected paediatric ophthalmic screening, triage, and surveillance pathways, particularly when combined with image-quality control, explicit escalation, and human oversight. However, confidence in the reported performance is limited by single-centre studies and enriched samples, small numbers of clinically important cases, heterogeneous analytical units, potentially optimistic aggregation procedures, limited external or temporal validation, incomplete calibration, and absent fairness analyses. Routine autonomous implementation remains premature.
