Skip Ribbon Commands
Skip to main content

Research: Improving OCT assessment in early glaucoma

Glaucoma is the leading cause of irreversible blindness globally, with a global prevalence of 3.5% in persons aged 40 to 80. The number of people with glaucoma worldwide is estimated to increase from 76 to 112 million between 2020 and 2040. In glaucoma, there is progressive loss of retinal ganglion cells and axons within the optic nerve, leading to structural and functional damage and eventual visual impairment, and blindness. The gradual loss of axons results in the thinning of the retinal nerve fiber layer (RNFL).

It is possible to measure the RNFL damage using optical coherence tomography (OCT) imaging of the optic nerve. However, the ability to accurately diagnose early glaucoma remains inadequate due to the considerable overlap of structural thickness measurements between normal and glaucoma eyes. Current commercial OCT systems only account for basic demographics (e.g. age) and not for individual differences such as disc area and glasses power. Incorporation of this data into the analysis could help improve the usability and reliability of OCT especially for early disease detection.

In the cross-sectional study, lead by Prof Leopold Schmetterer, our researchers developed a novel multivariate normative database of RNFL thickness by incorporating multiple demographic and anatomic factors into the model. This has resulted in a much improved glaucoma discrimination capability compared than conventional comparisons; and suggests that demographic and anatomic variance such as ethnicity, age, refractive error, optic disc (area, orientation, and ratio), fovea (distance and angle) as well as retinal vessel density) are all important in assessing RNFL thickness in our patients. This new normative database enables us to better assess patients, allowing for early detection and treatment if glaucoma is present, and avoiding un-necessary treatment of normal patients.

We show two healthy patients without glaucoma who benefited from using the compensation approach.

Figure 1 shows a 56-year-old healthy participant presenting abnormal RNFL probability color codes in the global average, superior and temporal quadrant with Cirrus normative database. However, this patient also had a small disc. In conventional OCT systems, the RNFL will usually be measured in a circumpapillary scanning circle of 3.4mm around the disc, and in a small disc, this would be in a more peripheral (or thinner) zone. After adjusting for disc size, RNFL was found to be in the normal range.

Figure 2 shows a 61-year-old healthy participant presenting abnormal RNFL probability color codes in the superior and inferior quadrants with Cirrus normative database. However, the patient also had high myopia, and in conventional OCT systems, this will involve a larger area, as the eyeball is longer. This often results in ‘thinner’ RNFL especially in the superior and inferior poles. Such changes can be misattributed to glaucoma especially when patients also have suspicious visual field defects.

Image and paper source:

For more information, watch this video as Dr Jacqueline Chua discusses on her research work:
Internet access required


Contributed by:


 Prof Leopold Schmetterer
 Scientific Director and Head 
 Ocular Imaging Research Group
 Singapore Eye Research Institute


 Dr Jacqueline Chua
 Junior Principal Investigator
 Ocular Imaging Research Group
 Singapore Eye Research Institute 

Back to EyeSight main page.