August 16, 2016
Diabetic Retinopathy is a complication of the eye that can affect anyone who has diabetes either type I or type II. It causes damage to the blood vessels that supply the retina – the seeing part of the eye. Worse, it lacks early symptoms making early detection essential.
Diabetic Retinopathy (“DR”) is considered one of the main causes of visual impairment. The World Health Organization considers it a priority eye disease.
Currently, detection is based on screening of the retina. This is a straightforward procedure that uses the assessment of visual acuity and eye fundus photography. Then a medical specialist, who looks for signs of diabetic retinopathy, grades the results for use in further diagnosis. Diagnosis is therefore heavily dependent on the quality of the images.
The iPRI Eye-Light project seeks to greatly enhance the screening of Diabetic Retinopathy by increasing the quality of these images.
A significant number of digital eye fundus images have strong contrast variations which can be a limiting factor for the diagnosis of the diabetic retinopathy lesions. Currently, to address this problem, graders have to manually adjust the image contrast.
- Current methods are not easily reproducible.
- Some images are considered un-gradable because they are simply too bright or too dark.
Causes of Ungradable Images
- Motion of the patient
- Acquisition conditions and differences in absorptions of the light in the eye
We have developed a fully automatic method, which achieves contrast uniformity across the entire image.
Our method is consistent with the physical principles of image creation. The contrast of dark or bright elements are adjusted in a way that emulate the real-world colour aspect of lesions such as micro-aneurysms or to anatomical structures such as veins. This method is much more accurate, useful and reliable than currently used grey level methods that use polynomial adjustment, mathematical morphology or histogram equalisations.
Results to Date
It has been tested on thousands of images acquired from different screening services ranging from a centre with quality controlled process while others were obtained under harsher conditions.
Some images were bright while others were dark making diagnosis difficult. However for all images, the lighting variations have been corrected and the contrast has been enhanced for lesions such as micro-aneurysms and the vascular structures. They are now easier to be detected by graders.
This new colour contrast method is a very promising tool. It helps graders diagnose the presence of diabetic retinopathy and other lesions present in digital eye fundus images. There is no other method that makes lesions more evident.
Our method also has the capacity to be rolled out widely and quickly. It is fully automated and it is easily integrated into a screening workflow.
iPRI, with its extensive expertise in global public health, epidemiology and translatable science into practical public health solutions, has identified a need for delivering a practical solution to the DR diagnosis challenge. iPRI’s international network of Senior Research Fellows, including recognized leaders in the study and treatment of diabetes, have confirmed this need and validated iPRI’s approach.
Armed with this support, iPRI believes that its software development investment will enhance the long-term quality of colour, eye fundus exams both in high- and low-resource countries. So far, tests on relevant databases have revealed that DR experts show interest.
This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 717108.