Feasibility of Multispectral Imaging for Retinal Imaging
Research type
Research Study
Full title
Feasibility of a multispectral macular imaging (MMI) system for detecting and monitoring morphological features of the macula
IRAS ID
315782
Contact name
Tariq Aslam
Contact email
Sponsor organisation
University of Manchester
Duration of Study in the UK
0 years, 7 months, 30 days
Research summary
Research Summary:
This is a proof-of-concept study where we will assess the feasibility of using images from a relatively inexpensive camera, that detects separate wavelengths of normal light, to determine information about the physical structures of the eye. This information is normally obtained only through much more expensive hospital laser-based equipment. If this is achievable it would open the door to a much more accessible and affordable system of monitoring patient disease that patients could even have in their own homes. \nThe part of the retina (back of the eye) that is most important to measure is the central area, the macula. All of the patients in the study will be under hospital care for disease of the macula and be receiving the complex and expensive laser scans as part of their usual care, that measure thickness of the macula accurately. \nFor this study we will use a purpose-built camera that uses different wavelengths of normal light to also image their eyes. Our objective is to then see if we can apply applying image processing techniques to the images we obtain from this normal light camera to develop a measure of macula thickness that is comparable to the gold standard laser scan. \nThere will be two stages to the study. Stage 1 is the stage for the assessment and development of the imaging and image analysis (MMI) system. (1 month, up to 20 patients). For this phase, we will assess and optimise the use of the MMI camera and analysis system by comparing images and developing algorithms to convert the data in them to a thickness score which will be as close as possible to that measured by the gold standard laser scans. \nStage 2 is the actual feasibility study where we will calculate and report the accuracy of the developed MMI system to determine both macular thickness and changes in macular thickness (Over 6 months, 40 patients with one repeat visit) \nUltimately, we hope to be able to demonstrate a system for measuring macular thickness and changes in macular thickness using a potentially cheap and accessible system. Both of these measures would be very useful for home monitoring of eye disease in future.Summary of Results:
Unfortunately, we were not able to assess the multispectral macular imaging camera system as it was not functional. Alternatively, we assessed patients using software to extract colour channels from white light images (fundus cameras) to produce intensity maps reflecting one colour channel minus another. Optical coherence tomography provided images for the gold standard assessment of retinal thickness. We looked at the feasibility and possible utility of hospital cameras in providing some evidence in monitoring physical macular structure and changes using image analysis from a fundus camera. Looking at the data there was no trend, statistical, or clinical relevance in using hospital cameras to monitor macular structure and changes. There was no significant correlation found. We also examined the values for the correlation coefficient and found a weak positive correlation between central subfield thickness and the index of the green–red component. Additionally, there was a weak correlation between central subfield thickness and both the green–blue and red–green indices. It was concluded that the use of hospital cameras with natural light and removing certain components from light to assess the macula and monitor macular changes, whether progression or regression, is neither feasible nor effective.REC name
South Central - Oxford B Research Ethics Committee
REC reference
22/SC/0232
Date of REC Opinion
2 Aug 2022
REC opinion
Further Information Favourable Opinion