Image

Performance of Large Language Models for Structured Recognition and Refractive Prediction

Performance of Large Language Models for Structured Recognition and Refractive Prediction

Recruiting
18 years and older
All
Phase N/A

Powered by AI

Overview

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.

Description

This study compares three large language models accessed in their native configurations, without fine-tuning or external tools. For each examination, the original IOLMaster 700 report image was supplied without manual annotation or pre-processing. A standardized instruction required: (i) structured extraction of AL, ACD, LT, WTW, K1/K2 and axes, ΔK, TK1/TK2 and axes, and ΔTK; (ii) binary toric candidacy and T-code according to institutional ALCON mapping; and (iii) refractive recommendations (sphere, cylinder, implantation axis). Each model generated three independent outputs per case. De-identification and IRB oversight (waiver of consent) were implemented according to institutional policy. The unit of enrollment is participants (n=54), with outcomes analyzed per eye (162 eyes) and per model generation where applicable.

Eligibility

Inclusion Criteria:

-postoperative corrected distance visual acuity (CDVA) of 0.10 logMAR or better -an absolute IOL rotational stability of less than 10∘ at the 1-month follow-up examination

Exclusion Criteria:

  • incomplete biometric data on the examination report;
  • a history of previous ocular surgery or ocular trauma
  • the occurrence of intraoperative complications, such as an anterior capsular tear or posterior capsular rupture
  • the development of significant postoperative complications, including but not limited to severe intraocular infection or inadequate pupillary dilation.

Study details
    Cataract

NCT07183891

Jin Yang

15 October 2025

Step 1 Get in touch with the nearest study center
We have submitted the contact information you provided to the research team at {{SITE_NAME}}. A copy of the message has been sent to your email for your records.
Would you like to be notified about other trials? Sign up for Patient Notification Services.
Sign up

Send a message

Enter your contact details to connect with study team

Investigator Avatar

Primary Contact

  Other languages supported:

First name*
Last name*
Email*
Phone number*
Other language

FAQs

Learn more about clinical trials

What is a clinical trial?

A clinical trial is a study designed to test specific interventions or treatments' effectiveness and safety, paving the way for new, innovative healthcare solutions.

Why should I take part in a clinical trial?

Participating in a clinical trial provides early access to potentially effective treatments and directly contributes to the healthcare advancements that benefit us all.

How long does a clinical trial take place?

The duration of clinical trials varies. Some trials last weeks, some years, depending on the phase and intention of the trial.

Do I get compensated for taking part in clinical trials?

Compensation varies per trial. Some offer payment or reimbursement for time and travel, while others may not.

How safe are clinical trials?

Clinical trials follow strict ethical guidelines and protocols to safeguard participants' health. They are closely monitored and safety reviewed regularly.
Add a private note
  • abc Select a piece of text.
  • Add notes visible only to you.
  • Send it to people through a passcode protected link.