BioE MS Thesis Presentation- Vidisha Goyal

Ross C. Ethier, Ph.D. (Advisor) (Department of Biomedical Engineering, Georgia Institute of Technology and Emory University)
Brandon Dixon, Ph.D. (School of Mechanical Engineering, Georgia Institute of Technology)
Pamela T. Bhatti, Ph.D.  (School of Electrical and Computer Engineering, Georgia Institute of Technology)

 

Deep Learning-based Optic Nerve Analysis 

Axon loss and degeneration are used to quantify the progression of several neurodegenerative diseases such as glaucoma, multiple sclerosis, etc. in animal models. In glaucoma, the gold standard for quantifying nerve health post-mortem is manual counting of axons from light micrographs of the optic nerve, which is subjective and laborious. This research is focused on developing a deep-learning model to segment normal-appearing axons, their axoplasm, and myelin sheath, from whole optic nerve images. These segmentation maps are fed into an image-processing pipeline for post-processing and for computing morphometric properties such as axoplasmic area, eccentricity, diameter, etc. With this technology, we will be able to answer important questions such as “Which axon size is preferentially damaged during glaucoma?” and “How does axon morphology change with increase optic nerve damage?” etc. Therefore, a reference RGC axonal atlas for Brown Norway rats was also constructed. A reference atlas of optic nerve RGC axonal morphological metrics could facilitate studies of neuro-ophthalmic diseases, such as glaucoma, by allowing sensitive detection of subtle RGC axonal changes and help answer some of the questions posed above.