LLM-Generated Lay Summaries for Brain MRI Reports

NCT07310394 | Enrolling By Invitation

• 1 other identifier: 2025-3783
• Study Type: interventional
• Target: 1,200
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this clinical trial is to learn if a summary written by artificial intelligence (AI) helps adults understand brain MRI reports for headaches. The main question it aims to answer is: "Does adding a simple summary help readers correctly understand if a cause for the headache was found in the report?" Researchers will compare standard MRI reports to reports that include an AI-generated explanation to see if the extra summary improves understanding. Participants will: Read 6 fictional brain MRI reports online. Answer questions to check if they understood the results. Rate their satisfaction and if they feel they would need to ask a doctor for help.

Conditions

Headache

Keywords

headache; LLM; Large Language Model; MRI; lay summary

Outcome Measures

Primary Outcomes (1)

• Objective understanding of the report

  Assessment of the participant's ability to correctly understand the medical findings. After reading each of the six fictional brain MRI reports, participants answer the binary question: "Is a probable explanation for the headache found in this report?" (Yes/No). The outcome is calculated as the proportion of correct responses compared to the ground truth of the specific clinical scenario (normal, incidental finding, or explanatory abnormality).

  Through completion of each response, an average of 5 minutes

Secondary Outcomes (3)

• Self-reported understanding score

  Through completion of each response, an average of 5 minutes

• Perceived need for professional clarification

  Through completion of each response, an average of 5 minutes

• Perceived ability to explain results to a relative

  Through completion of each response, an average of 5 minutes

Study Arms (2)

Original MRI reports

NO INTERVENTION

LLM-generated summaries in addition to the original reports

EXPERIMENTAL

Other: LLM-generated lay summary

Interventions

LLM-generated lay summary OTHER

Participants assigned to this group read fictional brain MRI reports that include an additional summary paragraph generated by an artificial intelligence tool. Specifically, an open-weights Large Language Model (LLM) with fewer than 100 billion parameters is used, hosted locally on a secure server to ensure data privacy. This model generates a short synthesis designed to be clear and structured for non-medical readers. This summary is inserted into the report under the heading Synthesis intended for the patient and non-radiologist physician. The intervention consists solely of this added text; the standard medical content of the report remains unchanged.

LLM-generated summaries in addition to the original reports

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Participants of the COMPARE e-cohort

You may not qualify if:

• None

Sponsors & Collaborators

• Direction Générale de l'Offre de Soins: collaborator
• University Hospital, Lille: lead

Study Sites (1)

Lille University Hospital

Lille, 59000, France

Location

MeSH Terms

Conditions

Headache

Condition Hierarchy (Ancestors)

Pain; Neurologic Manifestations; Signs and Symptoms; Pathological Conditions, Signs and Symptoms

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: OTHER
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Head of the Include Health Data Warehouse, Lille university Hospital

Study Record Dates

• First Submitted: November 28, 2025
• First Posted: December 30, 2025
• Study Start: March 25, 2026
• Primary Completion: April 30, 2026
• Study Completion: April 30, 2026
• Last Updated: May 4, 2026

Record last verified: 2026-03

Locations

FR (1)