BG-CNN: A Boundary Guided Convolutional Neural Network for Corneal Layer Segmentation from Optical Coherence Tomography

Precise segmentation of corneal layers depicted on optical coherence tomography (OCT) images plays an important role in detecting corneal diseases, such as keratoconus and dry eye. In this study, we present a boundary guided convolutional neural network (BG-CNN) to extract different corneal layers. The developed network uses three convolutional blocks to construct two network modules derived from the classical U-Net network. This network was trained based on a dataset consisting of 1712 OCT images. The segmentation results demonstrated the developed network achieved an average dice similarity coefficient (DSC) of 0.9599 and an interFiguresection over union (IOU) of 0.9253 on an independent testing set from our dataset, and it outperformed some available segmentation networks.

[1]  Jian Zhang,et al.  A coarse-to-fine deep learning framework for optic disc segmentation in fundus images , 2019, Biomed. Signal Process. Control..

[2]  Bram van Ginneken,et al.  A survey on deep learning in medical image analysis , 2017, Medical Image Anal..

[3]  Xiaochun Cao,et al.  Joint Optic Disc and Cup Segmentation Based on Multi-Label Deep Network and Polar Transformation , 2018, IEEE Transactions on Medical Imaging.

[4]  Terry Kim,et al.  Anatomy and physiology of the cornea , 2011, Journal of cataract and refractive surgery.

[5]  Jiss Kuruvilla,et al.  A review on image processing and image segmentation , 2016, 2016 International Conference on Data Mining and Advanced Computing (SAPIENCE).

[6]  Lei Wang,et al.  Simultaneous segmentation and bias field estimation using local fitted images , 2018, Pattern Recognit..

[7]  William J Feuer,et al.  The use of Bowman's layer vertical topographic thickness map in the diagnosis of keratoconus. , 2014, Ophthalmology.

[8]  Daniel Rueckert,et al.  Deep Learning for Cardiac Image Segmentation: A Review , 2020, Frontiers in Cardiovascular Medicine.

[9]  Xiaodong Yang,et al.  Level set based segmentation using local fitted images and inhomogeneity entropy , 2020, Signal Process..

[10]  Ryan P. McNabb,et al.  Topography and pachymetry maps for mouse corneas using optical coherence tomography. , 2019, Experimental eye research.

[11]  Joseph A. Izatt,et al.  Robust automatic segmentation of corneal layer boundaries in SDOCT images using graph theory and dynamic programming , 2011, Biomedical optics express.

[12]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.

[13]  Mohamed Abdel-Mottaleb,et al.  Automatic Segmentation of Corneal Microlayers on Optical Coherence Tomography Images , 2019, Translational vision science & technology.

[14]  Lei Wang,et al.  SDFN: Segmentation-based Deep Fusion Network for Thoracic Disease Classification in Chest X-ray Images , 2018, Comput. Medical Imaging Graph..

[15]  Mohamed Abou Shousha,et al.  Diagnosis and management of conjunctival and corneal intraepithelial neoplasia using ultra high-resolution optical coherence tomography. , 2011, Ophthalmology.

[16]  Hossein Rabbani,et al.  Optical Coherence Tomography Image Analysis , 2016 .

[17]  Chi-Wing Fu,et al.  Patch-Based Output Space Adversarial Learning for Joint Optic Disc and Cup Segmentation , 2019, IEEE Transactions on Medical Imaging.

[18]  Lei Wang,et al.  Active contours driven by edge entropy fitting energy for image segmentation , 2018, Signal Process..

[19]  Dexi Zhu,et al.  Ultra-high resolution optical coherence tomography for imaging the anterior segment of the eye. , 2011, Ophthalmic surgery, lasers & imaging : the official journal of the International Society for Imaging in the Eye.

[20]  Jian Zhang,et al.  Automated segmentation of the optic disc using the deep learning , 2019, Medical Imaging: Image Processing.

[21]  David A. Clausi,et al.  A Novel Algorithm for Extraction of the Layers of the Cornea , 2009, 2009 Canadian Conference on Computer and Robot Vision.

[22]  C. V. Jiji,et al.  Two stage contour evolution for automatic segmentation of choroid and cornea in OCT images , 2019, Biocybernetics and Biomedical Engineering.

[23]  Lei Wang,et al.  An active contour model based on local fitted images for image segmentation , 2017, Inf. Sci..