Glaucoma is a leading cause of irreversible blindness worldwide, with diagnosis and monitoring often limited by subjectivity and variability in conventional methods. Artificial intelligence (AI) has emerged as a transformative tool in ophthalmology, offering potential to improve diagnostic accuracy, predict disease progression, and guide surgical decision-making. A systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, with protocol registration in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD420261278225). PubMed, Embase, Scopus, and Cochrane Library were searched from January 2010 to December 2025 using predefined terms related to glaucoma and AI. Eligible studies included original research evaluating AI algorithms for glaucoma detection, progression monitoring, risk stratification, treatment prediction, or surgical decision support. Data extraction captured study design, population, AI methodology, imaging modality, and performance metrics. Quality assessment was performed using the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool for diagnostic studies and qualitative appraisal for prognostic and workflow studies. A random-effects meta-analysis was conducted for diagnostic accuracy studies reporting the area under the curve (AUC). Thirteen studies were included in the qualitative synthesis, with a subset contributing to the quantitative meta-analysis of diagnostic accuracy, encompassing over 75,000 images and more than 10,000 patients across retrospective, prospective, longitudinal, and multicenter cohorts. AI applications are clustered into three domains: diagnosis, progression forecasting, and surgical planning. Diagnostic studies demonstrated consistently high accuracy, with pooled meta-analysis confirming a summary AUC of 0.93 (95%CI: 0.91-0.95). Multimodal approaches achieved the highest performance (AUC 0.95), outperforming single-modality fundus or optical coherence tomography systems. AI models were shown to reliably predict visual field deterioration years in advance, enabling earlier intervention. There was evidence of AI’s role in surgical planning, including risk stratification, candidate selection, and structural imaging for pre-surgical assessment. Workflow integration, demonstrating the feasibility of embedding AI decision support into routine practice, was also discussed. AI has established itself as a powerful tool for glaucoma diagnosis, progression forecasting, and surgical decision-making. Quantitative synthesis confirms specialist-level diagnostic accuracy, while narrative evidence highlights emerging applications in risk stratification and clinical workflow integration. Future research should prioritise prospective multicenter validation, explainable AI frameworks, and randomised controlled trials to ensure that these technologies translate into meaningful improvements in surgical planning and long-term vision preservation.
Verma et al. (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: