Agentic Artificial Intelligence in Eye Banking: A Proposed Workflow

Area of Science:

• Ophthalmology
• Artificial Intelligence
• Medical Informatics

Background:

• Eye banking operations often involve fragmented manual workflows.
• Standardization and efficiency are critical for optimal tissue utilization in corneal transplantation.

Purpose of the Study:

• To develop an agentic artificial intelligence (AI) framework for streamlining and standardizing eye bank operations.
• Automate donor screening, corneal image analysis, and tissue suitability assessment under expert supervision.

Main Methods:

• Designed a modular, on-premises, multi-agent system with specialized AI agents.
• Agents perform donor screening, multimodal corneal image analysis, and integrated suitability assessment.
• Local operation ensures data privacy, real-time processing, and regulatory compliance.

Main Results:

• The AI framework replaces manual workflows with a unified, intelligent system, reducing variability and accelerating evaluations.
• Consistent and rapid analysis of donor data and corneal images improves tissue utilization and decreases discard rates.
• A modular, low-latency design facilitates deployment in diverse settings, including resource-limited regions.

Conclusions:

• Agentic AI provides a scalable solution for standardizing eye banking through automation and human oversight.
• This approach enhances decision quality, operational efficiency, and equity in corneal transplantation.
• The framework has implications for both limited-resource settings and transplantation medicine globally.