Multimodal 3-D segmentation of optic nerve head structures from spectral domain Oct volumes and color fundus photographs

Standard structural methods for the assessment of glaucoma, such as the planimetry of stereo color photographs of the optic disc, involve a subjective component either by the patient or examiner and suffer from poor reproducibility. Spectral-domain optical coherence tomography (SD-OCT) provides a 3-D, cross-sectional, microscale depiction of biological tissues. Given the wealth of volumetric information at microscale resolution available with SD-OCT volumes of the back of the eye, it is likely that better parameters can be obtained for measuring glaucoma changes that move beyond what is possible using fundus photography. The neural canal opening (NCO) is a 3-D single anatomic structure in SD-OCT volumes. It is proposed as a basis for a stable reference plane from which various optic nerve morphometric parameters can be derived. The overall aim of this Ph.D. project is to develop a framework to segment the 3-D NCO and retinal vessels using information from SD-OCT volumes and fundus photographs to aid in the management of glaucoma. Based on the mutual positional relationship of the NCO and vessels, a multimodal 3-D scale-learning-based framework is developed to iteratively identify these structures in SD-OCT volumes by incorporating each other’s pre-identified positional information. The algorithm first applies a 3-D wavelet-transform-learning-based layer segmentation and pre-segments the NCO using a graph-theoretic approach. To improve the NCO segmentation, the vessels are identified either using an OCT vessel segmentation approach by incorporating the pre-segmented NCO positional informa-