Performance of Large Language Models for Structured Recognition and Refractive Prediction

Sponsor
Jin Yang
Study ID
NCT07183891
Status
Recruiting
Apply to this trial

Conditions

  • Cataract

Eligibility Criteria

Sex
ALL
Age
18 Years - N/A
Healthy Volunteers
Not accepted

Study Details

We conducted a single-center, retrospective observational study to evaluate large language models (ChatGPT 4o, GPT-5, DeepSeek) for automated interpretation of de-identified IOLMaster 700 reports provided as raster images. Models produced structured biometric extraction, toric IOL recommendation, and refractive predictions (sphere, cylinder, axis). Primary outcomes included parameter-level agreement and refractive error metrics; secondary outcomes included decision-support performance for toric IOL selection and agreement on ordered T-codes. No clinical intervention was performed.

Key Dates

First listed
Sep 19, 2025
Start date
Aug 1, 2025
Status verified
Sep 2025
Primary completion
Dec 31, 2030
Completion
Dec 31, 2035

Study Design

Enrollment
100 participants (estimated)

Primary Outcome Measure

Refractive prediction error for sphere [ Time Frame: At index examination ]

Central Contacts

Related Studies