Because of the ever increasing size of OCT(A) data sets, manual segmentation of ocular layers is typically impractically laborious, necessitating the usage of automatic algorithms. While automatic segmentation of ocular layers in the normal eye can be challenging, automatic segmentation in the presence of pathology, which often causes significant ocular distortions, is more challenging still.

Perhaps nowhere is the demand for accurate segmentation more stringent than in OCTA analysis of the choriocapillaris. Analysis of the choriocapillaris is of great interest in studying a number of diseases, including age-related macular degeneration (AMD) and diabetic retinopathy (DR), where the choriocapillaris is thought to play an important role. En face OCTA analysis of the choriocapillaris typically requires segmentation of Bruch’s membrane (BM), an acellular matrix situated between the retinal pigment epithelium (RPE) and choriocapillaris.

However, to our knowledge, all previous publications on segmenting the BM have exclusively used OCT data. We think that OCTA data can be used synergistically alongside OCT data to segment Bruch’s membrane, as well as other ocular structures.

As such, the scope of this project is to explore different ways of incorporating and exploiting the additional information provided by the OCTA data in order to get more accurate and robust segmentation results for the BM in the presence of pathology.

This project is carried out in collaboration with Prof. Fujimoto's Biomedical Optical Imaging and Biophotonics Group at MIT in Cambridge, USA.