An AI solution to evaluate thyroid eye disease activity
Research type
Research Study
Full title
Development of an artificial intelligence solution to evaluate the activity and severity of thyroid eye disease using digital facial images.
IRAS ID
356967
Contact name
Vickie Lee
Contact email
Sponsor organisation
THYROSCOPE INC.
Clinicaltrials.gov Identifier
n/a, n/a
Duration of Study in the UK
1 years, 5 months, 31 days
Research summary
Why?
This study explores whether an AI system can accurately assess thyroid eye disease (TED) severity using digital facial images. TED, common in Graves' disease, can cause irreversible eye damage. An AI tool could enable early detection and self-assessment, leading to timely specialist care.What?
The study focuses on TED severity assessment using AI. The system will analyze digital eye images to predict TED activity, exophthalmos (eye protrusion), and palpebral fissure shape. It is a retrospective study comparing AI predictions to ophthalmologists’ evaluations.Who?
Eligible participants are adults (19+) suspected or diagnosed with TED. Excluded are those with facial deformities or prior eye surgery. The study will include 400-500 participants.Where?
The study is conducted at Imperial College Healthcare NHS Trust (Western Eye Hospital, London), in collaboration with THYROSCOPE INC. (South Korea).How?
Over 12 months, researchers will:Collect Data: De-identified digital eye images and medical records from TED patients.
Develop AI Models: Train and validate AI using medical records as a reference.
Compare Results: Assess AI accuracy against expert evaluations.
Expected Outcome
A validated AI tool for TED self-assessment, enabling earlier detection and medical intervention via smartphone images.REC name
West Midlands - Solihull Research Ethics Committee
REC reference
26/WM/0002
Date of REC Opinion
21 Jan 2026
REC opinion
Favourable Opinion