Artificial Intelligence in Neuro-Ophthalmology: Opportunities for the Diagnosis of Optic Neuropathies and Visual Pathway Disorders

Artificial intelligence (AI) is increasingly transforming the landscape of neuro-ophthalmology by enabling earlier and more precise identification of optic nerve and visual pathway disorders. With the growing complexity of multimodal diagnostic imaging and functional assessments, AI offers a scalable solution to enhance diagnostic accuracy and streamline clinical workflows. Recent advancements, particularly in deep learning (DL) and convolutional neural networks (CNNs), have shown notable potential in interpreting fundus photography, optical coherence tomography (OCT), and magnetic resonance imaging (MRI), facilitating the detection of conditions such as optic neuritis (ON), ischemic optic neuropathy, papilledema, and glaucomatous optic nerve damage. In parallel, AI-driven analysis of visual field (VF) tests has demonstrated improved consistency in assessing disease progression, supporting longitudinal monitoring. The development of mobile diagnostic applications and integrated decision-support systems further extends the utility of AI, particularly in settings with limited specialist access. Despite these promising innovations, critical challenges remain. These include data heterogeneity across populations and imaging platforms, the opaque nature of many AI models, which limits clinical interpretability, and the absence of standardized regulatory and ethical guidelines. As the field moves toward broader clinical adoption, success will depend on robust multicenter validation studies, the creation of explainable AI (XAI) frameworks, and the implementation of strong governance structures to ensure safety, fairness, and accountability in patient care.