DRUNET: a dilated-residual U-Net deep learning network to segment optic nerve head tissues in optical coherence tomography images.

Sripad Krishna Devalla,Giridhar Subramanian,Tin Aung,Liang Zhang,Nicholas G Strouthidis,Michaël J A Girard,Khai Sing Chin,Tin A Tun,Prajwal K Renukanand,Alexandre H Thiéry,Shamira Perera,Jean-Martial Mari,Bharathwaj K Sreedhar

doi:10.1364/boe.9.003244

Abstract

Given that the neural and connective tissues of the optic nerve head (ONH) exhibit complex morphological changes with the development and progression of glaucoma, their simultaneous isolation from optical coherence tomography (OCT) images may be of great interest for the clinical diagnosis and management of this pathology. A deep learning algorithm (custom U-NET) was designed and trained to segment 6 ONH tissue layers by capturing both the local (tissue texture) and contextual information (spatial arrangement of tissues). The overall Dice coefficient (mean of all tissues) was 0.91 ± 0.05 when assessed against manual segmentations performed by an expert observer. Further, we automatically extracted six clinically relevant neural and connective tissue structural parameters from the segmented tissues. We offer here a robust segmentation framework that could also be extended to the 3D segmentation of the ONH tissues.

Highlights

The development and progression of glaucoma is characterized by complex 3D structural changes within the optic nerve head (ONH) tissues
These include the thinning of the retinal nerve fiber layer (RNFL) [1,2,3]; changes in the minimum-rim-width [4], choroidal thickness [5, 6], lamina cribrosa (LC) depth [7,8,9], and posterior scleral thickness [10]; and migration of the LC insertion sites [11, 12]
The cohort consisted of normal controls, subjects with primary open angle glaucoma (POAG) and 19 subjects with primary angle closure glaucoma (PACG)

Summary

Introduction

The development and progression of glaucoma is characterized by complex 3D structural changes within the optic nerve head (ONH) tissues These include the thinning of the retinal nerve fiber layer (RNFL) [1,2,3]; changes in the minimum-rim-width [4], choroidal thickness [5, 6], lamina cribrosa (LC) depth [7,8,9], and posterior scleral thickness [10]; and migration of the LC insertion sites [11, 12]. While there exist several traditional image processing tools to automatically segment the ONH tissues [13,14,15,16,17,18,19,20], and extract these parameters, each tissue currently requires a different algorithm (tissue-specific). In our previous study [26], while it was possible to isolate the connective and neural tissues of the ONH, we were unable to segment each ONH tissue separately

Objectives

Methods

Results

Conclusion