Innovation in Diabetic Retinopathy Imaging

Innovation in Diabetic Retinopathy Imaging

The past few years has seen several advances in the management of diabetic eyes diseases, particularly in the diagnosis and treatment of diabetic retinopathy (DR) and diabetic macular edema (DME). Some of the key areas of ongoing research are summarised below.

New imaging techniques

The standard method for the detection of DR and DME is a comprehensive dilated eye examination with stereoscopic colour fundus photographs taken over time in order to monitor changes in the retina. 

Other imaging techniques, such as optical coherence tomography (OCT) and fluorescein angiography also play a role in the diagnosis and management of complications of DR, particularly for diabetic macular edema (DME). 

OCT is non-invasive technique that uses light to capture detailed 3-dimensional images showing the structure of the retina making it more suitable for the detection of DME than fundus photography, which only provides 2-dimensional images. OCT is commonly used in clinical practice to confirm the diagnosis and monitor the treatment of DME. However, its role as a screening test for DR (in addition to fundus photography) is still unclear, in part due to the low availability of OCT machines and personnel to operate them. [1] Several newer versions of OCT have been introduced including: 

  • Spectral-domain-OCT (SD-OCT): provides detailed images of morphological abnormalities in the retina; this approach is already widely used in clinical practice 
  • OCT-Angiography: this a relatively new approach allowing detailed visualisation of blood vessels in the retina and choroid layer of the eye, which can help monitor DR progression. An advantage of OCT-angiography is that its non-invasive and does not require injection of drugs to help visualise the blood vessel, which is required for fluorescein angiography) [2]
  • Swept source-OCT (SS-OCT): captures images of structures across a wider area than is reached with SD-OCT (including vitreous, retina, choroid, and sclera), and offers faster scan speeds

Other imaging modalities that may play a more prominent role in the management of DR and DME in the future, include:

  • Ultra-Widefield Color Fundus Photography: Compared with traditional single-field fundus photography, ultra-widefield imaging can capture higher quality images from a larger area of the retina, including the peripheral retina (the part that provides our side vision), which is often affected by DR. Studies have shown that this approach improves diagnosis and classification of DR[3]. The availability of ultra-widefield cameras has led to the increased use of ultra-wide field fluorescein angiography, a technique which allows better visualisation of the blood vessels in the peripheral retina,  improving the detection and classification of diabetic retinopathy.[4]
  • Retinal oximetry is a non-invasive technique that can be used to measure the oxygen saturation of retinal blood vessels. DR is caused by damage to the retinal blood vessels and the reduced blood supply can lead to a deficit of oxygen (hypoxia) in the inner retina. This oxygen deficit has been linked to the development of DR, and also to high oxygen levels in the retinal blood vessels. Therefore, measurement of higher oxygen saturation in blood vessels  through the use of retinal oximetry may in the future provide a method of detecting the early development of DR before it appears in other imaging modalities[5]
  • Fundus Autofluorescence is a non-invasive imaging technique that uses the fluorescent properties of lipofuscin, a naturally occurring compound found within the cells of retina to generate high-contrast retinal images that can be used in the diagnosis and management of several retinal disorders. Fundus autofluorescence has been shown to compare favourably to OCT and fluorescein angiography for the detection of macular oedema, and is likely to play a role in the diagnosis and follow-up of DR in the future.

Technologies to improve the implementation of screening 

Other technological advances that are likely to increase the implementation of screening of DR in the future include the availability of portable fundus cameras, and the use of smartphone-based camera for the capture of retinal images. [6]These advances will be particularly useful in rural areas of developing countries when incorporated into a tele-screening programme.

The incorporation of automated DR image assessment systems that help to eliminate the requirement of manual (human) experts to assess and grade images are also likely to have a significant impact. These systems have demonstrated the potential to reduce the cost of screening. Several different automated grading platforms have been developed over the past few years, and some of the more recent platforms employ artificial-intelligence software, one of which was recently approved by the US FDA (https://www.eyediagnosis.net/idx-dr)