Morphological segmentation and level classification for diabetic and hypertensive retinopathy through ophthalmic images and convolutional networks
Article Sidebar
Main Article Content
The main objective of this research is to perform segmentation and classification of diabetes and hypertensive retinopathy retinal fundus images. Using a combination of a convolutional network UNet and a ConvNet was proposed for vessel mask segmentation and retinopathy classification, respectively. The classification process relies on ten defined classes, where values ranging from 0 to 4 represent diabetic retinopathy, and values ranging from 5 to 9 correspond to hypertensive retinopathy. The approximate results in the segmentation were Jaccard index of 74%, F1 of 85%, and an Accuracy of 96%, and in the classification was Accuracy of 80%.
OPS. Organización Panamericana de la Salud. (Online). (cited 2023 09 23. Available from: https://www.paho.org/es/temas/enfermedades-no-transmisibles.
Manresa JM, Forés R, Vásquez X, Alzamora MT, Heras A, Delgado P, et al. Fiabilidad de la retinografía para la detección de retinopatía hipertensiva en Atención Primaria. Atención Primaria. 2020 Jun; 52(6). DOI: https://doi.org/10.1016/j.aprim.2019.06.005
Centro de Oftalmología Bonafonte. Assessment of Corneal Angiography Filling Patterns in Corneal Neovascularization. Journal of Clinical Medicine. 2023; 12(2). DOI: https://doi.org/10.3390/jcm12020633
Poříz V. Diabetic Retinopathy Detection Using Neural Networks. Tesis. Czech Technical University in Prague; 2020.
Jha A,VA&AAR. Association of severity of diabetic retinopathy with corneal endothelial and thickness changes in patients with diabetes mellitus. Eye. 2022 June; 36(6). DOI: https://doi.org/10.1038/s41433-021-01606-x
Zhou J CB. Retinal Cell Damage in Diabetic Retinopathy. Cells. 2023 May; 12(9). DOI: https://doi.org/10.3390/cells12091342
Jiménez Jiménez L. Manifestaciones oftalmológicas vasculares de la hipertensión arterial. Tesis. Universidad de Sevilla, Medicina; 2020.
Hagos MT. Point-of-Care Diabetic Retinopathy Diagnosis: A Standalone Mobile Application Approach. Tesis. Johns Hopkins University; 2020.
Jiménes García J. Detección de vasos sanguíneos en retinografías mediante técnicas de procesado digital de imágenes. Tesis. Valladolid: Universidad de Valladolid, Escuela Técnica Superior de Ingenieros de Telecomunicación; 2017.
Kropp M,GO,MAea. Diabetic retinopathy as the leading cause of blindness and early predictor of cascading complications—risks and mitigation. EPMA Journal. 2023 March 01; 14(1).
Wu S MX. Optic Nerve Regeneration in Diabetic Retinopathy: Potentials and Challenges Ahead. Int J Mol Sci. 2023 January 11; 24(2). DOI: https://doi.org/10.3390/ijms24021447
Jiménez Jiménez L. Manifestaciones Oftalmológicas Vasculares de la Hipertensión Arterial. Trabajo de Grado. Sevilla: Universidad de Sevilla, Departamento de Medicina; 2020.
Cardona Suárez JC, Fernández Agudelo FN. Modelo de Machine Learning para clasificación de pacientes con glaucoma en la población del Valle del Cauca. Tesis. Santiago de Cali: Universidad Icesi, Ingeniería; 2022.
Unnati V S, Tripathy K. Diabetic Retinopathy: StatPerls; 2023.
Sarki R. Automatic Detection Of Diabetic Eye Disease Through Deep Learning Using Fundus Images. Tesis. Victoria University; 2021. DOI: https://doi.org/10.1109/ACCESS.2020.3015258
Russell G. Diabetic retinal imaging: methods in automatic processing. Tesis. The University of Manchester; 2014.
Kropp M, Golubnitschaja O, Mazurakova A, Koklesova L, Sargheini N, Vo TTKS, et al. Diabetic retinopathy as the leading cause of blindness and early predictor of cascading complications—risks and mitigation. EPMA Journal. 2023 January; 14(1). DOI: https://doi.org/10.1007/s13167-023-00314-8
Abbas Q, Ibrahim M. DenseHyper: an automatic recognition system for detection of hypertensive retinopathy using dense features transform and deep-residual learning. Multimedia Tools and Applications. 2020 November; 79. DOI: https://doi.org/10.1007/s11042-020-09630-x
Cortés Rodríguez DC. Cámara retinal: herramienta de telediagnóstico para detección de retinopatía diabética, retinopatía hipertensiva y glaucoma. Tésis. Universidad de la Salle; 2016.
Merchán Barrezueta MJ, Lucas Baño ES, Sánchez Escobar DA, Arellano Blacio MA. Retinopatía diabética e hipertensiva. RECIAMUC. 2023 January; 7(1): p. 290-298.
Chala M, Nsiri B, Hachem M, Soulaymani A, Mokhtari A, Benaji B. An automatic retinal vessel segmentation approach based on Convolutional Neural Networks. Expert Systems with Applications. 2021; 184: p. 115459. DOI: https://doi.org/10.1016/j.eswa.2021.115459
Zhang H, Qiu Y, Song C, Li J. Landmark-Assisted Anatomy-Sensitive Retinal Vessel Segmentation Network. Diagnostics. 2023; 13(13). DOI: https://doi.org/10.3390/diagnostics13132260
NA S, Yadav AK, Akbar M, Kumar M, Yadav D. Retinal blood vessel segmentation using a deep learning method based on modified U-NET model. 2021 September. DOI: https://doi.org/10.36227/techrxiv.16653238
Phong Thanh Nguyen VDBHKDVPTPEYGPJ. An Optimal Deep Learning Based Computer-Aided Diagnosis System for Diabetic Retinopathy. Computers, Materials & Continua. 2021; 61(3): p. 2815-2830. DOI: https://doi.org/10.32604/cmc.2021.012315
Takahashi HaTHaAYaIYaKH. Applying artificial intelligence to disease staging: Deep learning for improved staging of diabetic retinopathy. PLoS ONE. 2017; 12(6). DOI: https://doi.org/10.1371/journal.pone.0179790
Dutta S&MB&BM&CR&INCSN. Classification of Diabetic Retinopathy Images by Using Deep Learning Models. International Journal of Grid and Distributed Computing. 2018; 11: p. 89-106. DOI: https://doi.org/10.14257/ijgdc.2018.11.1.09
Kaggle. Diabetic Retinopathy Detection. (Online).; 2015 (cited 2023 April 10. Available from: https://www.kaggle.com/c/diabetic-retinopathy-detection/data.
Drive. DRIVE (Digital Retinal Images for Vessel Extraction). (Online). (cited 2022 06 22. Available from: https://paperswithcode.com/dataset/drive.
Fire. Kaggle. (Online).; 2020 (cited 2022 03 13. Available from: https://www.kaggle.com/datasets/andrewmvd/fundus-image-registration.
Mendeley Data. Data on Fundus Images for Vessels Segmentation, Detection of Hypertensive Retinopathy, Diabetic Retinopathy and Papilledema. (Online).; 2020 (cited 2023 04 10. Available from: https://data.mendeley.com/datasets/3csr652p9y/2.
Kaggle. Ocular Disease Recognition. (Online).; 2020 (cited 2022 February 10. Available from: https://www.kaggle.com/datasets/andrewmvd/ocular-disease-recognition-odir5k.
Singh A. CLAHE augmentation -Ranzcr comp. (Online).; 2020 (cited 2022 11 15. Available from: https://www.kaggle.com/code/amritpal333/clahe-augmentation-ranzcr-comp.
Manizheh Safarkhani G, Hojjat M, Mehdi A. Segmentation of Retinal Blood Vessels Using U-Net++ Architecture and Disease Prediction. Electronics. 2022; 11(21). DOI: https://doi.org/10.3390/electronics11213516
Verdeguer Gómez J. Redes neuronales para la clasificación y segmentación de imágenes médicas. Tesis. Valencia: Universitat Politècnica de València, Departamento de Sistemas Informáticos y Comput; 2020.
Ronneberger O,FP,BT. U-Net: Convolutional Networks for Biomedical Image Segmentation. Medical Image Computing and Computer-Assisted Intervention. 2015 November 18; 9351. DOI: https://doi.org/10.1007/978-3-319-24574-4_28
Tomar N. Github. (Online).; 2021 (cited 2022 04 20. Available from: https://github.com/nikhilroxtomar/Retina-Blood-Vessel-Segmentation-in-PyTorch/tree/main/UNET.
He K, Zhang X, Ren S, Sun J. Deep Residual Learning for Image Recognition. 2015. DOI: https://doi.org/10.1109/CVPR.2016.90
Zhang Y, Liu S, Li C, Wang J. Rethinking the Dice Loss for Deep Learning Lesion Segmentation in Medical Images. Journal of Shanghai Jiaotong University (Science). 2021 Febrary; 26(1): p. 93. DOI: https://doi.org/10.1007/s12204-021-2264-x
Wazir S, Moazam Fraz M. HistoSeg: Quick attention with multi-loss function for multi-structure segmentation in digital histology images. 2022 12th International Conference on Pattern Recognition Systems ICPRS. 2022 June. DOI: https://doi.org/10.1109/ICPRS54038.2022.9854067
Zhuang L, Hanzi M, Chao-Yuan W, Christoph F, Trevor D, Saining X. A ConvNet for the 2020s. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2022;: p. 11966-11976.
Persson A. aladdinpersson/Machine-Learning-Collection. (Online).; 2020 (cited 2022 06 20. Available from: https://github.com/aladdinpersson/Machine-Learning-Collection.
Ferreras Extremo A. Estudio de algoritmos de redes neuronales convolucionales en dataset de imágenes médicas. Valladolid: Universidad de Valladolid; 2021.
Agarwal V. Complete Architectural Details of all EfficientNet Models. (Online).; 2020 (cited 2023 April 28. Available from: https://towardsdatascience.com/complete-architectural-details-of-all-efficientnet-models-5fd5b736142.
Yasoda BY, Jagannadham DBV. Classification of Brain Tumor Using Finetuned Efficientnet. International Journal of Creative Research Thoughts (IJCRT). 2022 December; 10(12).
Marques G, Ferreras A, de la Torre-Diez I. An ensemble-based approach for automated medical diagnosis of malaria using EfficientNet. Multimed Tools Appl. 2022 March 29; 81(19). DOI: https://doi.org/10.1007/s11042-022-12624-6
Alhichri H&AA&BY&AN&AN. Classification of Remote Sensing Images Using EfficientNet-B3 CNN Model with Attention. IEEE Access. 2021 January. DOI: https://doi.org/10.1109/ACCESS.2021.3051085
Amreen B, Yung-Cheol B. Lightweight EfficientNetB3 Model Based on Depthwise Separable Convolutions for Enhancing Classification of Leukemia White Blood Cell Images. IEEE Access. 2023; 11: p. 37203-37215. DOI: https://doi.org/10.1109/ACCESS.2023.3266511
Lasloum T, Alhichri H, Bazi Y, Alajlan N. SSDAN: Multi-Source Semi-Supervised Domain Adaptation Network for Remote Sensing Scene Classification. Remote Sensing. 2021 September; 13. DOI: https://doi.org/10.3390/rs13193861
Tan M, V. Le Q. EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks Chaudhuri KaSR, editor.; 2019.
PyTorch. PyTorch Linear Layer (Fully Connected Layer) Explained. (Online).; 2022 (cited 2023 April 12. Available from: https://androidkt.com/pytorch-linear-layer-fully-connected-layer-explained/.
Moczulski M, Denil M, Appleyard J, Freitas N. ACDC: A Structured Efficient Linear Layer. 2016 May.
Kumar B. PyTorch Batch Normalization. (Online).; 2022 (cited 2023 April 12. Available from: https://pythonguides.com/pytorch-batch-normalization/.
Educba. PyTorch Dropout. (Online).; 2022 (cited 2023 April 12. Available from: https://www.educba.com/pytorch-dropout.
Buslaev A, Parinov A, Khvedchenya E, Iglovikov V, Kalinin A. Albumentations: fast and flexible image augmentations. Computer Science > Computer Vision and Pattern Recognition. 2018 Sep; 11(2). DOI: https://doi.org/10.3390/info11020125
Gothwal R, Gupta S, Gupta D, Kumar A. Color image segmentation algorithm based on RGB channels. Proceedings of 3rd International Conference on Reliability, Infocom Technologies and Optimization. 2014;: p. 1-5. DOI: https://doi.org/10.1109/ICRITO.2014.7014669
Shouting Feng ZZDPQT. CcNet: A cross-connected convolutional network for segmenting retinal vessels using multi-scale features. Neurocomputing. 2020; 392: p. 268-276. DOI: https://doi.org/10.1016/j.neucom.2018.10.098
Baiyuan D, Gongjian W, Conghui M, Xiaoliang Y. Target recognition in synthetic aperture radar images using binary morphological operations. Journal of Applied Remote Sensing. 2016; 10(4). DOI: https://doi.org/10.1117/1.JRS.10.046006
SH K, JY K. Application of Fast Non-Local Means Algorithm for Noise Reduction Using Separable Color Channels in Light Microscopy Images. Int J Environ Res Public Health. 2021; 18(6). DOI: https://doi.org/10.3390/ijerph18062903
Mohammad H, Sulaiman MN. A Review on Evaluation Metrics for Data Classification Evaluations. International Journal of Data Mining & Knowledge Management Process. 2015 Mar; 5. DOI: https://doi.org/10.5121/ijdkp.2015.5201
Ogwok D, Ehlers EM. Jaccard Index in Ensemble Image Segmentation: An Approach. CIIS '22: Proceedings of the 2022 5th International Conference on Computational Intelligence and Intelligent. 2022 November;: p. 9-14. DOI: https://doi.org/10.1145/3581792.3581794
Qin J, Liu T, Wang Z, Liu L, Fang H. GCT-UNET: U-Net Image Segmentation Model for a Small Sample of Adherent Bone Marrow Cells Based on a Gated Channel Transform Modul. Electronics. 2022; 11(22). DOI: https://doi.org/10.3390/electronics11223755
Xu S, Chen Z, Cao W, Zhang F, Tao B. Retinal Vessel Segmentation Algorithm Based on Residual Convolution Neural Network. Frontiers in Bioengineering and Biotechnology. 2021; 9. DOI: https://doi.org/10.3389/fbioe.2021.786425
- Francisco Bernal Baquero, Darwin E. Martínez, Human action detection for inventory control using computer vision , Ingeniería y Competitividad: Vol. 26 No. 1 (2024): Ingeniería y Competitividad
Accepted 2023-11-22
Published 2024-02-26
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors grant the journal and Universidad del Valle the economic rights over accepted manuscripts, but may make any reuse they deem appropriate for professional, educational, academic or scientific reasons, in accordance with the terms of the license granted by the journal to all its articles.
Articles will be published under the Creative Commons 4.0 BY-NC-SA licence (Attribution-NonCommercial-ShareAlike).