Artificial intelligence in ophthalmic optical coherence tomography angiography: a bibliometric analysis.
Abstract
Aim
To analyze research landscapes, hotspots, and trends in artificial intelligence (AI) applied to ophthalmic optical coherence tomography angiography (OCTA) from 2015 to 2024, aiming to reveal global collaborations and guide future research.
Methods
We retrieved publications from Web of Science Core Collection using AI and OCTA-related search terms. After screening, 237 articles were analyzed with CiteSpace (version 6.3.R1) for co-citation, collaboration, and keyword burst analysis.
Results
Annual publications grew rapidly since 2017. China (94 publications) and the US (86 publications) were the most productive and influential countries. Research focused on: deep learning algorithms and vessel segmentation, ocular disease diagnosis including diabetic retinopathy and glaucoma, and automated vascular quantification and multimodal imaging. Key limitations included small sample sizes and limited algorithm generalizability.
Conclusion
The integration of AI with OCTA has significantly advanced the automation and precise quantitative analysis of ophthalmic vascular imaging, demonstrating substantial potential for diagnosing various retinal and choroidal diseases. Future efforts should focus on developing more robust and interpretable AI models, conducting large-scale, multi-center clinical validation, and establishing standardized imaging and analysis protocols to ultimately realize the widespread adoption of AI-OCTA technology in routine clinical practice.