NCT07328776

An Assistant Model for Anti-VEGF Therapy Decision

A Multi-stage Evaluation Foundation Model for Anti-VEGF Therapy Decision: A Randomized Control Trial

• Status: Not Yet Recruiting
• Phase: N/A
• Study type: Interventional
• Enrollment: 160 (estimated)
• Sponsor: Shanghai General Hospital, Shanghai Jiao Tong University School of Medicine · Academic / Other
• Sex: All
• Age: 50 Years – 85 Years
• Healthy volunteers: Not accepted

Summary

We present FM-VEGF-CDSS system, the first foundation model-based AI system specifically designed for the decision of anti-VEGF therapy by emulating clinician decision-making reasoning process for those with exudative retinal diseases. FM-VEGF-CDSS is capable of processing multimodal input including optical coherence tomography (OCT) and descriptive medical metadata. Trained and validated on cases from centers across China, the system performed well in both internal and external validation dataset, and could generate the convenient report to enhance decision-making interpretability. To validate its clinical practice, we conducted a prospective multicenter RCT, rigorously assessing FM-VEGF-CDSS system's prediction accuracy and real-world utility. This will improve the standardized decision for anti-VEGF therapy, promoting health equity.

Conditions

• Exudative Age-Related Macular Degeneration
• DME
• Retinal Vein Occlusion (RVO)

Interventions

Type	Name	Description
OTHER	FM-VEGF-CDSS assisted	A Comprehensive Deep Learning Model for Assisting the decision of anti-VEGF therapy: FM-VEGF-CDSS system
OTHER	without FM-VEGF-CDSS	without FM-VEGF-CDSS

Timeline

• Start date: 2026-01-01
• Primary completion: 2026-08-31
• Completion: 2026-08-31
• First posted: 2026-01-09
• Last updated: 2026-01-09

Locations

2 sites across 1 country: China