Current Applications of Artificial Intelligence in Neuro-Ophthalmic Imaging: A Narrative Approach
Medical Sciences · 2026
DOI: 10.3390/medsci14040422 ↗Kurz gefasst
Übersichtsarbeit in der Zeitschrift Medical Sciences, 2026. Verfasst von Raluca Eugenia Iorga und 5 weiteren Autor:innen. Beteiligte Einrichtung: Grigore T. Popa University of Medicine and Pharmacy, University Street No 16, 700115 Iasi, Romania. Hauptthemen laut MEDLINE: Artificial Intelligence, Ophthalmology. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 159 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
The eye provides a valuable window for the identification of neurological diseases, as pathological changes may involve both the retina and the optic nerve. Standard imaging of the retina and the optic nerve are non-invasive and are useful tools in investigating the integrity of the visual pathways. With population aging and the increasing burden of neuro-ophthalmic diseases, there is a growing need for tools that can assist clinicians in achieving rapid and reliable diagnoses. AI may facilitate the detection of specific neuro-ophthalmological conditions.
Methods
This article presents a narrative review of previously published studies addressing the current applications of AI in selected neuro-ophthalmic conditions, with emphasis on their advantages, limitations, and challenges in screening, diagnosis, and disease progression assessment.
Results
AI-based approaches provide valuable opportunities for the screening, characterization, and monitoring of optic nerve head abnormalities in neuro-ophthalmology. Recent advancements, particularly in deep learning and convolutional neural networks, have shown notable potential in interpreting fundus photography, optical coherence tomography and magnetic resonance imaging, supporting a comprehensive assessment of optic nerve structure and function. These systems are particularly promising because they can extract quantitative information from routine imaging modalities and may assist clinicians in detecting structural damage that is difficult to identify through conventional evaluation alone.
Conclusions
AI-assisted ophthalmic imaging has shown promising diagnostic performance in identifying ONH abnormalities and retinal or optic nerve changes associated with neurological disease. At present, however, most applications remain within research, retrospective validation, or proof-of-concept settings, and evidence that AI consistently improves early diagnosis or clinical outcomes beyond conventional assessment remains limited. Accordingly, AI should currently be regarded as a complementary tool that may support screening, image interpretation, disease characterization, and referral decisions rather than replace established clinical evaluation.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Grigore T. Popa University of Medicine and Pharmacy, University Street No 16, 700115 Iasi, Romania.
- Department of Gastroenterology, "L. Pasteur" Clinical Hospital, 28630 Le Coudray, France.
ORCID-Profile: Vlad Constantin Donica, Ciprian Danielescu, Camelia Margareta Bogdănici
Zitieren
APA
Iorga, R. E., Donica, V. C., Danielescu, C., Bogdănici, C. M., Munteanu-Dănulescu, R. S., & Moraru, A. D. (2026). Current Applications of Artificial Intelligence in Neuro-Ophthalmic Imaging: A Narrative Approach. Medical Sciences. https://doi.org/10.3390/medsci14040422
Vancouver
Iorga RE, Donica VC, Danielescu C, Bogdănici CM, Munteanu-Dănulescu RS, Moraru AD. Current Applications of Artificial Intelligence in Neuro-Ophthalmic Imaging: A Narrative Approach. Medical Sciences. 2026. doi:10.3390/medsci14040422
BibTeX
@article{Iorga2026,
author = {Iorga, Raluca Eugenia and Donica, Vlad Constantin and Danielescu, Ciprian and Bogdănici, Camelia Margareta and Munteanu-Dănulescu, Răzvana Sorina and Moraru, Andreea Dana},
title = {Current Applications of Artificial Intelligence in Neuro-Ophthalmic Imaging: A Narrative Approach},
journal = {Medical Sciences},
year = {2026},
doi = {10.3390/medsci14040422}
}
EUVIMED