Mapping the Intelligence: A Bibliometric Analysis of Artificial Intelligence and Machine Learning Applications in Retinal OCT for Neurodegenerative Diseases.
Abstract
Background
Retinal optical coherence tomography (OCT) combined with artificial intelligence (AI) is increasingly proposed as a scalable window onto neurodegenerative disease, yet the intellectual architecture of this field remains unmapped.
Methods
Following the BIBLIO guideline, 1,036 documents (1976-2026) were retrieved from the Web of Science Core Collection and analysed with bibliometrix (v5.4.0), integrating performance analysis, collaboration and co-citation networks, and science mapping.
Results
Output grew exponentially after 2010, peaking at 139 articles in 2025. Productivity and impact concentrated in a transatlantic core (USA, Germany, UK; Charité Berlin, University College London, Johns Hopkins; Calabresi, Paul, Saidha), with a 19-journal Bradford core led by Investigative Ophthalmology & Visual Science. Thematic mapping documented a decisive post-2022 reorientation toward deep learning, optical coherence tomography angiography (OCT-A) and oculomics, with AI themes consolidating as motor themes.
Conclusions
The field is transitioning from structural biomarker research to computational phenotyping; externally validated, globally inclusive collaborations will determine whether retinal AI fulfils its screening potential.