Early-stage Glaucoma screening platform using combination of Visual Field Testing and Fundus imaging mechanisms

Glaucoma is an irreversible neurodegenerative disease which mainly damages the optical nerve part of human eye progressively. Worldwide, nearly 60 million people are affected by glaucoma, where around 90% of these people are from the low- and middle-income countries. And this number is expected to raise above 111 million worldwide by 2040. The prevalence of glaucoma is very high in low-income countries like Ethiopia, compared with the developed countries. In Ethiopia, glaucoma is mentioned as one of the the public health challenges causing for vision loss even though the causes are highly preventable if treated at an early stage. The painlessness of the disease makes it to be unknown until it reaches an irreversible stage. Doing a screening test at an early stage will be an ideal solution to stop the disease from worsening. However, since symptoms are not noticed at the early level, people will not become aware of the situation. Clinically, visual field test, and optic nerve head assessments are the most effective procedures for early glaucoma screening. The limited number of professionals in the field of ophthalmology and optometry, and the inaccessibility and the expensiveness of medical devices for such examinations play a huge role for glaucoma to be a real threat for people who especially live in the rural side of Ethiopia, where the majority of the population of the country actually resides. Due to the limitation of these resources, visual field testing and/or ONH assessments are hardly conducted in the primary health sectors during glaucoma screening. Additionally, it is not that common to conduct subjective and objective examinations. Moreover, relying the final screening decision on only one of the assessment approaches may not give full information about the progress and occurrence of the disease. In this PhD work, we will develop an integrated platform that can provide a subjective output from the smartphone based visual field screening platform, and an objective output from the deep learning based glaucoma screening from fundus imaging approach. Having subjective and objective outputs will help the health professionals not to depend the final decision only on either of the two techniques, rather to give them a reliable decision. This platform will be developed and evaluated in collaboration with Jimma University and UZ Leuven, taking into consideration of the limited resources in Ethiopia. Deep learning techniques will be introduced as part of this automated solution for processing the fundus images that are taken from patients. And a smartphone based visual field screening platform will be used for assessing the subjective information about the disease. The combination of the two assessments from the patient will be investigated to see if it can boost screening of glaucoma at early stage. A clinical study will be set up with sufficient number of participants in selected healthcare centers in Belgium and in Ethiopia to test the validity of the new screening approach. Moreover, we will adopt this way of glaucoma screening in 2-3 health centers for validating the performance. One of the research questions is to quantify the added value of the combination of the two techniques. We will also address to what extent this approach can overcome the challenges of glaucoma screening campaigns in Ethiopia. The final output of this PhD research will be an affordable and easy-to-use platform/device for early glaucoma screening.