This research presents a novel RetSTA approach that leverages large language models to standardize clinical fundus image reports, addressing a critical challenge in healthcare data integration.

• Creates a bilingual standard terminology for fundus clinical terms
• Enables consistent diagnostic reporting across healthcare systems
• Improves the ability of AI models to understand diverse medical reports
• Demonstrates a practical application of LLMs in solving real-world healthcare challenges

Why it matters: Standardized medical reporting is essential for accurate diagnostics, effective patient care, and building robust healthcare databases that can power future AI innovations in medicine.

RetSTA: An LLM-Based Approach for Standardizing Clinical Fundus Image Reports