Active contour method for ILM segmentation in ONH volume scans in retinal OCT.

Kay Gawlik,Friedemann Paul,Frank Hausser,Alexander U Brandt,Ella Maria Kadas

doi:10.1364/boe.9.006497

Abstract

The optic nerve head (ONH) is affected by many neurodegenerative and autoimmune inflammatory conditions. Optical coherence tomography can acquire high-resolution 3D ONH scans. However, the ONH's complex anatomy and pathology make image segmentation challenging. This paper proposes a robust approach to segment the inner limiting membrane (ILM) in ONH volume scans based on an active contour method of Chan-Vese type, which can work in challenging topological structures. A local intensity fitting energy is added in order to handle very inhomogeneous image intensities. A suitable boundary potential is introduced to avoid structures belonging to outer retinal layers being detected as part of the segmentation. The average intensities in the inner and outer region are then rescaled locally to account for different brightness values occurring among the ONH center. The appropriate values for the parameters used in the complex computational model are found using an optimization based on the differential evolution algorithm. The evaluation of results showed that the proposed framework significantly improved segmentation results compared to the commercial solution.

Highlights

The eye’s retina is part of the central nervous system (CNS) and as such features a similar cellular composition as the brain [1]
Retinal alterations have been described in clinically isolated syndrome [2], multiple sclerosis [3], neuromyelitis optica spectrum disorders [4,5,6], Susac’s symdrome [7,8], Parkinson’s disease [9], and Alzheimer’s dementia [10]
We looked at the error contribution of the central region around the optic nerve head (ONH) consisting of all A-Scans within a radius of 1.5 mm

Summary

Introduction

The eye’s retina is part of the central nervous system (CNS) and as such features a similar cellular composition as the brain [1]. Many chronic brain conditions lead to retinal changes. The retina is the only part of the CNS that is readily accessible by optical imaging, putting great potential into its imaging in the context of these disorders. Spectral domain optical coherence tomography (SD-OCT) is the method of choice to acquire retinal 3D images in μm resolution [11, 12]. The human retina demonstrates two macroscopic landmarks, whose analysis is especially promising in the context of neurological disorders. Macular SD-OCT images can be analysed quantitatively with intra-retinal segmentation [13]. The derived thickness or volume changes can be used to quantify neuro-axonal damage in neurological disorders, e.g. in multiple sclerosis [14,15,16]

Methods

Results

Discussion

Conclusion