Automated detection of Hydroxychloroquine retinopathy
Research type
Research Study
Full title
Automated detection of Hydroxychloroquine retinopathy using optical coherence tomography and fundus autofluorescence
IRAS ID
276371
Contact name
Riaz Asaria
Contact email
Clinicaltrials.gov Identifier
Z6364106/2019/12/35, UCL data protection reference number
Duration of Study in the UK
1 years, 0 months, 29 days
Research summary
Hydroxychloroquine (HCQ) is an effective and commonly prescribed medication for a number of autoimmune conditions. Unfortunately, HCQ can also cause damage to the retina, known as HCQ retinopathy, resulting in irreversible central visual loss. Therefore, annual screening is recommended by the Royal College of Ophthalmologists (RCOphth). The recommended investigations for screening are a combination of optical coherence tomography (OCT) and fundus autofluorescence (FAF), as well as automated visual field testing. This currently takes place in hospital based ophthalmology clinics across the United Kingdom.
We plan to retrospectively and prospectively collect OCT and FAF images of patients currently undergoing screening for HCQ retinopathy at the Royal Free London NHS Foundation Trust and the Whittington Health NHS Trust. We will also analyse OCT images of patients taking HCQ from Heidelberg Engineering’s OCT database. Through classical and deep learning techniques, we aim to develop an algorithm that can interpret OCT and FAF images and automatically detect HCQ retinopathy. Ultimately this algorithm could be used to screen patients outside of busy ophthalmology clinics, decrease the number of images that currently require ophthalmologists to review, and may detect retinopathy at earlier stages and therefore allow for better visual outcomes.
REC name
London - Harrow Research Ethics Committee
REC reference
20/LO/1027
Date of REC Opinion
30 Oct 2020
REC opinion
Further Information Favourable Opinion