Segmentación de imágenes de OCT y OCT-A por medio de Redes Neuronales Convolucionales
DOI:
https://doi.org/10.17488/RMIB.43.3.2Palabras clave:
Segmentación OCT-A, ResU-Net, segmentación FCN, Red neuronal convolucionalResumen
La segmentación juega un papel vital en las imágenes de angiografía por tomografía de coherencia óptica (OCT-A), ya que la separación y distinción de las diferentes partes que forman la mácula simplifican la detección posterior de patrones/enfermedades observables en la retina. En este trabajo, llevamos a cabo una segmentación de imágenes multiclase donde se destacan las mejores características en los plexos apropiados al comparar diferentes arquitecturas de redes neuronales, incluidas U-Net, ResU-Net y FCN. Nos centramos en dos zonas críticas: la segmentación de la vasculatura retiniana (RV) y la zona avascular foveal (FAZ). La precisión para RV y FAZ en 316 imágenes OCT-A de la base de datos OCT-A 500 se obtuvo en 93.21 % y 92.59 %. Cuando se segmentó la FAZ en una clasificación binaria, con un 99.83% de precisión.
Descargas
Citas
Wons J, Pfau M, Wirth MA, Freiberg FJ, et al. Optical coherence tomography angiography of the foveal avascular zone in retinal vein occlusion. Ophthalmologica [Internet]. 2016; 235:195-202. Available from: https://doi.org/10.1159/000445482
Guo M, Zhao M, Cheong AMY, Dai H, et al. Automatic quantification of superficial foveal avascular zone in optical coherence tomography angiography implemented with deep learning. Vis Comput Ind Biomed Art [Internet]. 2019;2:21. Available from: https://doi.org/10.1186/s42492-019-0031-8
Leopold HA, Orchard J, Zelek JS, Lakshminarayanan V. PixelBNN: Augmenting the Pixelcnn with Batch Normalization and the Presentation of a Fast Architecture for Retinal Vessel Segmentation. J Imaging [Internet]. 2019;5(2):26. Available from: https://doi.org/10.3390/jimaging5020026
Zhang Y, Chung ACS. Deep Supervision with Additional Labels for Retinal Vessel Segmentation Task. In: Frangi A, Schnabel J, Davatzikos C, Alberola-López C, et al. (eds). Medical Image Computing and Computer Assisted Intervention- MICCAI 2018 [Internet]. Granada, Spain: Springer; 2018:83-91. Available from: https://doi.org/10.1007/978-3-030-00934-2_10
Xiao X, Lian S, Luo Z, Li S. Weighted Res-UNet for High-Quality Retina Vessel Segmentation. In: 2018 9th International Conference on Information Technology in Medicine and Education (ITME) [Internet]. Hangzhou, China: IEEE; 2018: 327-331. Available from: https://doi.org/10.1109/ITME.2018.00080
Son J, Park SJ, Jung KH. Towards Accurate Segmentation of Retinal Vessels and the Optic Disc in Fundoscopic Images with Generative Adversarial Networks. J Digit Imaging [Internet]. 2019;32(3):499–512. Available from: https://doi.org/10.1007/s10278-018-0126-3
Jayabalan GS, Bille JF. The Development of Adaptive Optics and Its Application in Ophthalmology. In: Bille J. (eds). High Resolution Imaging in Microscopy and Ophthalmology [Internet]. Cham, Switzerland: Springer; 2019: 339–358p.
Taher F, Kandil H, Mahmoud H, Mahmoud A, et al. A Comprehensive Review of Retinal Vascular and Optical Nerve Diseases Based on Optical Coherence Tomography Angiography. Appl Sci [Internet]. 2021;11(9):4158. Available from: https://doi.org/10.3390/app11094158
Spaide RF, Fujimoto JG, Waheed NK, Sadda SR, et al. Optical coherence tomography angiography. Vol. 64, Prog Retin Eye Res [Internet]. 2018;64:1–55. Available from: https://doi.org/10.1016/j.preteyeres.2017.11.003
de Carlo TE, Romano A, Waheed NK, Duker JS. A review of optical coherence tomography angiography (OCTA). Int J Retin Vitr [Internet]. 2015;1:5. Available from: https://doi.org/10.1186/s40942-015-0005-8
Wylęgała A, Teper S, Dobrowolski D, Wylęgała E. Optical coherence angiography: A review. Medicine [Internet]. 2016;95(41):e4907. Available from: https://doi.org/10.1097/MD.0000000000004907
Azzopardi G, Strisciuglio N, Vento M, Petkov N. Trainable COSFIRE filters for vessel delineation with application to retinal images. Med Image Anal [Internet]. 2015;19(1):46–57. Available from: https://doi.org/10.1016/j.media.2014.08.002
Lau QP, Lee L, Hsu W, Wong TY. Simultaneously Identifying All True Vessels from Segmented Retinal Images. IEEE Trans Biomed Eng [Internet]. 2013;60(7):1851-1858. Available from: https://doi.org/10.1109/tbme.2013.2243447
Ghazal M, Al Khalil Y, Alhalabi M, Fraiwan L, et al. 9-Early detection of diabetics using retinal OCT images. In: El-Baz AS, Suri JS (eds). Diabetes and Retinopathy [Internet]. United States: Elsevier; 2020. 173–204p. Available from: https://doi.org/10.1016/B978-0-12-817438-8.00009-2
Li W, Zhang Y, Ji Z, Xie K, et al. IPN-V2 and OCTA-500: Methodology and Dataset for Retinal Image Segmentation, distributed by Cornell University [Internet]. 2020. Available from:
https://doi.org/10.48550/arXiv.2012.07261
Bates NM, Tian J, Smiddy WE, Lee W-H, et al. Relationship between the morphology of the foveal avascular zone, retinal structure, and macular circulation in patients with diabetes mellitus. Sci Rep [Internet]. 2018;8:5355. Available from: https://doi.org/10.1038/s41598-018-23604-y
Ronneberger O, Fischer P, Brox T. U-net: Convolutional Networks for Biomedical Image Segmentation. In: Navab N, Hornegger J, Wells W, Frangi A (eds). Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015 [Internet]. Munich, Germany: Springer; 2015: 234-241. Available from: https://doi.org/10.1007/978-3-319-24574-4_28
Sappa LB, Okuwobi IP, Li M, Zhang Y, et al. RetFluidNet: Retinal Fluid Segmentation for SD-OCT Images Using Convolutional Neural Network. J Digit Imaging [Internet]. 2021;34(3):691–704. Available from: https://doi.org/10.1007/s10278-021-00459-w
Rasamoelina AD, Adjailia F, Sinčák P. A Review of Activation Function for Artificial Neural Network. In: 2020 IEEE 18th World Symposium on Applied Machine Intelligence and Informatics (SAMI) [Internet]. Herlany, Slovakia: IEEE; 2020: 281-286. Available from: https://doi.org/10.1109/SAMI48414.2020.9108717
Qi W, Wei M, Yang W, Xu C, et al. Automatic Mapping of Landslides by the ResU-Net. Remote Sens [Internet]. 2020;12(15):2487. Available from: https://doi.org/10.3390/rs12152487
Ioffe S, Szegedy C. Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift. In: ICML'15: Proceedings of the 32nd International Conference on International Conference on Machine Learning - Volume 37 [Internet]. Lille, France: JMLR; 2015: 448–456. Available from: https://dl.acm.org/doi/10.5555/3045118.3045167
He Y, Carass A, Liu Y, Jedynak BM, et al. Fully Convolutional Boundary Regression for Retina OCT Segmentation. In: Shen D, Liu T, Peters TM, Staib LH, et al. (eds). Medical Image Computing and Computer Assisted Intervention – MICCAI 2019. MICCAI 2019 [Internet]. Shenzhen, China: Springer; 2019: 120–128. Available from: https://doi.org/10.1007/978-3-030-32239-7_14
Liu X, Song L, Liu S, Zhang Y. A review of Deep-Learning-Based Medical Image Segmentation Methods. Sustainability [Internet]. 2021;13(3):1224. Available from: https://doi.org/10.3390/su13031224
Descargas
Publicado
Cómo citar
Número
Sección
Licencia
Derechos de autor 2022 Revista mexicana de ingeniería biomédica.
Esta obra está bajo una licencia internacional Creative Commons Atribución-NoComercial 4.0.
Una vez que el artículo es aceptado para su publicación en la RMIB, se les solicitará al autor principal o de correspondencia que revisen y firman las cartas de cesión de derechos correspondientes para llevar a cabo la autorización para la publicación del artículo. En dicho documento se autoriza a la RMIB a publicar, en cualquier medio sin limitaciones y sin ningún costo. Los autores pueden reutilizar partes del artículo en otros documentos y reproducir parte o la totalidad para su uso personal siempre que se haga referencia bibliográfica al RMIB. No obstante, todo tipo de publicación fuera de las publicaciones académicas del autor correspondiente o para otro tipo de trabajos derivados y publicados necesitaran de un permiso escrito de la RMIB.
