Automatic Segmentation of Choroid Layer Using Deep Learning on Spectral Domain Optical Coherence Tomography

The purpose of this article is to evaluate the accuracy of the optical coherence tomography (OCT) measurement of choroidal thickness in healthy eyes using a deep-learning method with the Mask R-CNN model. Thirty EDI-OCT of thirty patients were enrolled. A mask region-based convolutional neural network (Mask R-CNN) model composed of deep residual network (ResNet) and feature pyramid networks (FPNs) with standard convolution and fully connected heads for mask and box prediction, respectively, was used to automatically depict the choroid layer. The average choroidal thickness and subfoveal choroidal thickness were measured. The results of this study showed that ResNet 50 layers deep (R50) model and ResNet 101 layers deep (R101). R101 U R50 (OR model) demonstrated the best accuracy with an average error of 4.85 pixels and 4.86 pixels, respectively. The R101 ∩ R50 (AND model) took the least time with an average execution time of 4.6 s. Mask-RCNN models showed a good prediction rate of choroidal layer with accuracy rates of 90% and 89.9% for average choroidal thickness and average subfoveal choroidal thickness, respectively. In conclusion, the deep-learning method using the Mask-RCNN model provides a faster and accurate measurement of choroidal thickness. Comparing with manual delineation, it provides better effectiveness, which is feasible for clinical application and larger scale of research on choroid.

[1]  E. Agrón,et al.  CHOROIDAL THICKNESS AND VASCULARITY VARY WITH DISEASE SEVERITY AND SUBRETINAL DRUSENOID DEPOSIT PRESENCE IN NONADVANCED AGE-RELATED MACULAR DEGENERATION. , 2020, Retina.

[2]  Chong Wang,et al.  Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search. , 2017, Biomedical optics express.

[3]  M. Pellegrini,et al.  Choroidal Vascularity Index: An In-Depth Analysis of This Novel Optical Coherence Tomography Parameter , 2020, Journal of clinical medicine.

[4]  June-Goo Lee,et al.  Deep Learning in Medical Imaging: General Overview , 2017, Korean journal of radiology.

[5]  Changhee Han,et al.  Synthesizing Diverse Lung Nodules Wherever Massively: 3D Multi-Conditional GAN-Based CT Image Augmentation for Object Detection , 2019, 2019 International Conference on 3D Vision (3DV).

[6]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  M. Yaseri,et al.  Choroidal Thickness in Acute Non-arteritic Anterior Ischemic Optic Neuropathy , 2020, Journal of ophthalmic & vision research.

[8]  Zicheng Qiu,et al.  Choroid Segmentation of Retinal OCT Images Based on CNN Classifier and l2-lq Fitter , 2021, Comput. Math. Methods Medicine.

[9]  Josh Wallman,et al.  The multifunctional choroid , 2010, Progress in Retinal and Eye Research.

[10]  Justis P. Ehlers,et al.  Evaluation of choroidal thickness in retinitis pigmentosa using enhanced depth imaging optical coherence tomography , 2012, British Journal of Ophthalmology.

[11]  F. Mutlu,et al.  Comparison of peripapillary choroidal thickness measurements via spectral domain optical coherence tomography with and without enhanced depth imaging , 2016, Postgraduate medicine.

[12]  Yun-Mi Song,et al.  Measurable Range of Subfoveal Choroidal Thickness With Conventional Spectral Domain Optical Coherence Tomography , 2018, Translational Vision Science & Technology.

[13]  Kiran Kumar Vupparaboina,et al.  Choroidal imaging biomarkers. , 2019, Survey of ophthalmology.

[14]  Changhee Han,et al.  MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction , 2020, BMC Bioinformatics.

[15]  Cason B. Robbins,et al.  Choroidal Structural Analysis in Alzheimer's Disease, Mild Cognitive Impairment, and Cognitively Healthy Controls. , 2020, American journal of ophthalmology.

[16]  Kaiming He,et al.  Rethinking ImageNet Pre-Training , 2018, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[17]  Kaiming He,et al.  Feature Pyramid Networks for Object Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  J. Lee,et al.  Peripapillary Choroidal Thickness Change of Polypoidal Choroidal Vasculopathy after Anti-vascular Endothelial Growth Factor , 2017, Korean journal of ophthalmology : KJO.

[19]  R. Agrawal,et al.  Decrease in Choroidal Vascularity Index of Haller’s layer in diabetic eyes precedes retinopathy , 2020, BMJ open diabetes research & care.

[20]  David Alonso-Caneiro,et al.  Automatic segmentation of OCT retinal boundaries using recurrent neural networks and graph search. , 2018, Biomedical optics express.

[21]  R. Spaide,et al.  Enhanced depth imaging spectral-domain optical coherence tomography. , 2008, American journal of ophthalmology.

[22]  Wolfgang Drexler,et al.  Choroid, Haller's, and Sattler's layer thickness in intermediate age-related macular degeneration with and without fellow neovascular eyes. , 2014, Investigative ophthalmology & visual science.

[23]  Serkan Akkaya Macular and Peripapillary Choroidal Thickness in Patients With Keratoconus. , 2018, Ophthalmic surgery, lasers & imaging retina.

[24]  U. Chakravarthy,et al.  POLYPOIDAL CHOROIDAL VASCULOPATHY FEATURES VARY ACCORDING TO SUB-FOVEAL CHOROIDAL THICKNESS. , 2020, Retina.

[25]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[26]  L. Pablo,et al.  Peripapillary Choroidal Thickness Analysis Using Swept-Source Optical Coherence Tomography in Glaucoma Patients: A Broader Approach , 2017, Ophthalmic Research.

[27]  Mia K Markey,et al.  Quality assessment for spectral domain optical coherence tomography (OCT) images , 2009, BiOS.

[28]  H. Oku,et al.  Long-Term Follow-Up Changes of Central Choroidal Thickness Thinning after Repeated Anti-VEGF Therapy Injections in Patients with Central Retinal Vein Occlusion-Related Macular Edema with Systemic Hypertension , 2019, Ophthalmologica.

[29]  Ross B. Girshick,et al.  Mask R-CNN , 2017, 1703.06870.

[30]  W. Lee,et al.  Morphologic features of large choroidal vessel layer: age-related macular degeneration, polypoidal choroidal vasculopathy, and central serous chorioretinopathy , 2018, Graefe's Archive for Clinical and Experimental Ophthalmology.

[31]  Nicolas Greliche,et al.  Increased choroidal thickness in primary angle closure measured by swept-source optical coherence tomography in Caucasian population , 2019, International Ophthalmology.

[32]  Shanna C. Yeung,et al.  ¬CHOROIDAL THICKNESS IN PATIENTS WITH CARDIOVASCULAR DISEASE: A REVIEW. , 2020, Survey of ophthalmology.

[33]  Namkug Kim,et al.  Deep Learning in Medical Imaging , 2019, Neurospine.

[34]  Qiang Wu,et al.  Automatic Choroid Layer Segmentation from Optical Coherence Tomography Images Using Deep Learning , 2019, Scientific Reports.

[35]  Hideki Koizumi,et al.  Enhanced depth imaging optical coherence tomography. , 2011, Ophthalmic surgery, lasers & imaging : the official journal of the International Society for Imaging in the Eye.

[36]  David Alonso-Caneiro,et al.  Automatic choroidal segmentation in OCT images using supervised deep learning methods , 2019, Scientific Reports.

[37]  Georg Langs,et al.  f‐AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks , 2019, Medical Image Anal..

[38]  D. Mackey,et al.  Choroidal thickness in young adults and its association with visual acuity. , 2020, American journal of ophthalmology.