ISI Journals

Glaucoma remains a leading cause of irreversible blindness worldwide, necessitating the development of advanced predictive systems to monitor and manage disease progression. Traditional clinical assessments often rely on optical coherence tomography (OCT) imaging and visual field tests; however, their interpretability and predictive accuracy are limited by the complexity of disease phenotypes and inter-patient variability. Recent advancements in artificial intelligence have enabled new opportunities for integrating imaging biomarkers with computational models to enhance diagnostic precision and progression prediction. Within this framework, knowledge distillation has emerged as a promising paradigm, allowing the transfer of knowledge from complex, high-capacity models to smaller, efficient, and clinically deployable architectures. Such approaches not only enable real-time inference in clinical environments but also mitigate computational burdens, thus broadening accessibility in resource-limited healthcare systems. Equity in artificial intelligence is a growing concern in ophthalmology, as biases in training datasets and algorithmic decision-making can disproportionately affect underrepresented populations. To address these challenges, equity-enhanced modeling integrates fairness-aware optimization strategies into predictive pipelines, ensuring that performance remains robust across demographic groups defined by age, sex, ethnicity, and socioeconomic status. This paradigm shift recognizes that glaucoma is not only a medical condition but also a public health challenge requiring equitable technological interventions. By combining OCT imaging, deep learning, and fairness-focused knowledge distillation, researchers are constructing predictive systems that are both accurate and socially responsible. The integration of digital medicine principles into glaucoma research represents a transformation in how disease progression is studied, monitored, and treated. Digital biomarkers extracted from OCT imaging allow continuous monitoring and more granular disease stratification, while knowledge distillation ensures that such predictive frameworks remain interpretable, lightweight, and deployable across varied clinical infrastructures. These advances align with the broader goals of personalized medicine, where prediction and treatment plans are tailored to the unique characteristics of each patient. This article systematically examines the methodological advancements, practical applications, and fairness considerations of equity-enhanced glaucoma prediction using OCT and knowledge distillation. The introduction frames the clinical urgency and technological opportunities, the literature review synthesizes existing findings, and the methodology outlines the integration of fairness-enhanced distillation pipelines. Results are presented to highlight improvements in predictive accuracy, computational efficiency, and fairness metrics, followed by a conclusion emphasizing the broader implications for ophthalmology, digital medicine, and artificial intelligence governance. The work provides a critical foundation for the development of equitable, interpretable, and clinically relevant predictive tools in ophthalmic care.