Maculopathies, a diverse group of eye diseases affecting the central portion of the retina, the macula, are a leading cause of vision loss globally. Early detection and accurate diagnosis are critical for timely intervention and preventing irreversible vision loss. However, traditional methods such as manually detecting the disease are time-consuming, tiring, and often lack sensitivity and specificity, hindering optimal patient care. Here we delve into the applications of machine learning and other deep learning algorithms for analyzing retinal images including optical coherence tomography (OCT). We explore how machine learning algorithms can automatically extract subtle features and patterns from retinal images, enabling accurate differentiation between healthy and diseased maculas, as well as classifying specific maculopathy conditions. This work showcases the most common retinal diseases, provides an overview of the prevalent imaging modalities, and presents a critical evaluation of current deep-learning research for the detection and diagnosis of drusen, diabetic macular edema (DME), and choroidal neovascularization (CNV). For the model, we proposed a residual neural network (ResNet), a convolutional neural network that allows high performance in image recognition. The deep learning model is structured so that learning takes place on a hierarchical set of representations and to update the network weights more efficiently we used the Adam (adaptive moment estimation) optimization algorithm, which is an extension of the Stochastic Gradient Descent (SGD) and is used instead of classical SGD. Overall, all the methods represent a significant advancement in disease detection and diagnosis using OCTs, and have the great potential to assist healthcare professionals and clinicians in informed decision-making, the early intervention of disease, and personalized treatment optimization.

Keywords:Maculopathy diagnosis, Optical Coherence Tomography (OCT), Machine Learning, Drusen, DME, CNV, ResNet.

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Serener, A & Kaymak, S, (2018). Automated age-related macular degeneration and diabetic macular edema detection on OCT images using deep learning.
2. Ali M. Alqudah, (2019), AOCT-NET: A CNN automated classification of multiclass retinal diseases using spectral-D domain optical coherence tomography images.
3. Yuchen Du, Q. Chen, Y. Fan, J. Zhu, (2021). An Automatic identification of myopic maculopathy related imaging features in optic disc region via machine learning methods.
4. Ye, J. Wang, Y. Chen, Zhe Lv, S. He, (2021). Automatic Screening and Identifying Myopic Maculopathy on an Optical Coherence Tomography Images Using Deep Learning, pg. 3-8.
5. Du, S. Xie, Y. Fang, T. Igarashi-Yokoi (2021). A Deep Learning Approach for Automated Detection of Myopic Maculopathy and Pathologic Myopia in Fundus Images, pg. 1-7.
6. Chetan P, Ramesh M, Anupriya K. (2023). The Design of a Novel Detection Using KNN Classification Technique for Early Sign of Diabetic Maculopathy, pg.1-6.
7. Mariacristina P, Usha C, Stela V, (2022). The Update on Current and Future Management for Diabetic Maculopathy, pg. 3-7.
8. Vasile P, Ibrahim A, Andreas H, (2019). The Detection of Diabetic Retinopathy andMaculopathy in Eye Fundus Images Using Deep Learning and Image Augmentation, pg.6-8.
9. Felix G, Judith M, Caroline B, (2018). A Deep Learning Algorithm for the Prediction of Age- Related Eye Disease Study Severity, pg.3-9.
10. X He1, P Ren, Li Lu, X Tang, J Wang, (2022). The Development of a deep learning algorithm for myopic maculopathy classification based on OCT images using transfer learning, pg. 4- 6.
11. Fang L, Cunefare D, Wang C, Guymer RH, Li S, Farsiu S., (2017). An Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search. Biomed Opt Express.
12. CS Lee, Tyring AJ, NP Deruyter, Wu Y, Rokem A, Lee AY, (2017). The Deep learning based, automated segmentation of macular edema in optical coherence tomography.
13. Lee CS, Lee AY, Baughman DM, (2017). Deep learning is effective for the classification of an OCT images of normal versus age- related macular degeneration. Ophthalmol Retina.
14. Devalla SK, Renukanand PK, Sreedhar BK, Subramanian G, Zhang L, Perera S, et al., (2018). DRUNET: A dilated-residual U-Net deep learning network to segment the optic nerve head tissues in optical coherence tomography images.
15. Perdomo O, Rios H, Otalora S, Rodriguez FJ, Meriaudeau F, Muller H, et al., (2019). The Classification of diabetes-related retinal diseases using a deep learning approach in optical coherence tomography images.
16. Li Y, Feng W, Zhao X, Liu B, Zhang Y, Chi W, et al., (2020). The Development and validation of a deep learning system to screen vision-threatening conditions in high myopia using optical coherence tomography images.
17. Wei L, He W, Wang J, Zhang K, Du Y, Qi J, et al., (2021). Optical coherence tomography-based deep learning algorithm for visual acuity prediction of highly myopic eyes after cataract surgery.
18. Ye X, Wang J, Chen Y, Lv Z, He S, Mao J, et al., (2021). Automatic screening and identifying myopic maculopathy on optical coherence tomography images using deep learning. Pan SJ, Yang Q., IEEE, (2010). A survey on transfer learning

Maculopathies, a diverse group of eye diseases affecting the central portion of the retina, the macula, are a leading cause of vision loss globally. Early detection and accurate diagnosis are critical for timely intervention and preventing irreversible vision loss. However, traditional methods such as manually detecting the disease are time-consuming, tiring, and often lack sensitivity and specificity, hindering optimal patient care. Here we delve into the applications of machine learning and other deep learning algorithms for analyzing retinal images including optical coherence tomography (OCT). We explore how machine learning algorithms can automatically extract subtle features and patterns from retinal images, enabling accurate differentiation between healthy and diseased maculas, as well as classifying specific maculopathy conditions. This work showcases the most common retinal diseases, provides an overview of the prevalent imaging modalities, and presents a critical evaluation of current deep-learning research for the detection and diagnosis of drusen, diabetic macular edema (DME), and choroidal neovascularization (CNV). For the model, we proposed a residual neural network (ResNet), a convolutional neural network that allows high performance in image recognition. The deep learning model is structured so that learning takes place on a hierarchical set of representations and to update the network weights more efficiently we used the Adam (adaptive moment estimation) optimization algorithm, which is an extension of the Stochastic Gradient Descent (SGD) and is used instead of classical SGD. Overall, all the methods represent a significant advancement in disease detection and diagnosis using OCTs, and have the great potential to assist healthcare professionals and clinicians in informed decision-making, the early intervention of disease, and personalized treatment optimization.

Keywords:Maculopathy diagnosis, Optical Coherence Tomography (OCT), Machine Learning, Drusen, DME, CNV, ResNet.

🔒 This is a subscription article

Full text is available to subscribers and institutional members. Please choose an option below to access it.

1. Serener, A & Kaymak, S, (2018). Automated age-related macular degeneration and diabetic macular edema detection on OCT images using deep learning.
2. Ali M. Alqudah, (2019), AOCT-NET: A CNN automated classification of multiclass retinal diseases using spectral-D domain optical coherence tomography images.
3. Yuchen Du, Q. Chen, Y. Fan, J. Zhu, (2021). An Automatic identification of myopic maculopathy related imaging features in optic disc region via machine learning methods.
4. Ye, J. Wang, Y. Chen, Zhe Lv, S. He, (2021). Automatic Screening and Identifying Myopic Maculopathy on an Optical Coherence Tomography Images Using Deep Learning, pg. 3-8.
5. Du, S. Xie, Y. Fang, T. Igarashi-Yokoi (2021). A Deep Learning Approach for Automated Detection of Myopic Maculopathy and Pathologic Myopia in Fundus Images, pg. 1-7.
6. Chetan P, Ramesh M, Anupriya K. (2023). The Design of a Novel Detection Using KNN Classification Technique for Early Sign of Diabetic Maculopathy, pg.1-6.
7. Mariacristina P, Usha C, Stela V, (2022). The Update on Current and Future Management for Diabetic Maculopathy, pg. 3-7.
8. Vasile P, Ibrahim A, Andreas H, (2019). The Detection of Diabetic Retinopathy andMaculopathy in Eye Fundus Images Using Deep Learning and Image Augmentation, pg.6-8.
9. Felix G, Judith M, Caroline B, (2018). A Deep Learning Algorithm for the Prediction of Age- Related Eye Disease Study Severity, pg.3-9.
10. X He1, P Ren, Li Lu, X Tang, J Wang, (2022). The Development of a deep learning algorithm for myopic maculopathy classification based on OCT images using transfer learning, pg. 4- 6.
11. Fang L, Cunefare D, Wang C, Guymer RH, Li S, Farsiu S., (2017). An Automatic segmentation of nine retinal layer boundaries in OCT images of non-exudative AMD patients using deep learning and graph search. Biomed Opt Express.
12. CS Lee, Tyring AJ, NP Deruyter, Wu Y, Rokem A, Lee AY, (2017). The Deep learning based, automated segmentation of macular edema in optical coherence tomography.
13. Lee CS, Lee AY, Baughman DM, (2017). Deep learning is effective for the classification of an OCT images of normal versus age- related macular degeneration. Ophthalmol Retina.
14. Devalla SK, Renukanand PK, Sreedhar BK, Subramanian G, Zhang L, Perera S, et al., (2018). DRUNET: A dilated-residual U-Net deep learning network to segment the optic nerve head tissues in optical coherence tomography images.
15. Perdomo O, Rios H, Otalora S, Rodriguez FJ, Meriaudeau F, Muller H, et al., (2019). The Classification of diabetes-related retinal diseases using a deep learning approach in optical coherence tomography images.
16. Li Y, Feng W, Zhao X, Liu B, Zhang Y, Chi W, et al., (2020). The Development and validation of a deep learning system to screen vision-threatening conditions in high myopia using optical coherence tomography images.
17. Wei L, He W, Wang J, Zhang K, Du Y, Qi J, et al., (2021). Optical coherence tomography-based deep learning algorithm for visual acuity prediction of highly myopic eyes after cataract surgery.
18. Ye X, Wang J, Chen Y, Lv Z, He S, Mao J, et al., (2021). Automatic screening and identifying myopic maculopathy on optical coherence tomography images using deep learning. Pan SJ, Yang Q., IEEE, (2010). A survey on transfer learning