A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images
Retinal Fluids (fluid collections) develop because of the accumulation of fluid in the retina, which may be caused by several retinal disorders, and can lead to loss of vision. Optical coherence tomography (OCT) provides non-invasive cross-sectional images of the retina and enables the visualization...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10043848/ |
| _version_ | 1797894952972189696 |
|---|---|
| author | Mohammad Rahil B. N. Anoop G. N. Girish Abhishek R. Kothari Shashidhar G. Koolagudi Jeny Rajan |
| author_facet | Mohammad Rahil B. N. Anoop G. N. Girish Abhishek R. Kothari Shashidhar G. Koolagudi Jeny Rajan |
| author_sort | Mohammad Rahil |
| collection | DOAJ |
| description | Retinal Fluids (fluid collections) develop because of the accumulation of fluid in the retina, which may be caused by several retinal disorders, and can lead to loss of vision. Optical coherence tomography (OCT) provides non-invasive cross-sectional images of the retina and enables the visualization of different retinal abnormalities. The identification and segmentation of retinal cysts from OCT scans is gaining immense attention since the manual analysis of OCT data is time consuming and requires an experienced ophthalmologist. Identification and categorization of the retinal cysts aids in establishing the pathophysiology of various retinal diseases, such as macular edema, diabetic macular edema, and age-related macular degeneration. Hence, an automatic algorithm for the segmentation and detection of retinal cysts would be of great value to the ophthalmologists. In this study, we have proposed a convolutional neural network-based deep ensemble architecture that can segment the three different types of retinal cysts from the retinal OCT images. The quantitative and qualitative performance of the model was evaluated using the publicly available RETOUCH challenge dataset. The proposed model outperformed the state-of-the-art methods, with an overall improvement of 1.8%. |
| first_indexed | 2024-04-10T07:18:45Z |
| format | Article |
| id | doaj.art-9d4b4d7563754102aaeb0b80de498cdf |
| institution | Directory Open Access Journal |
| issn | 2169-3536 |
| language | English |
| last_indexed | 2024-04-10T07:18:45Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj.art-9d4b4d7563754102aaeb0b80de498cdf2023-02-25T00:02:08ZengIEEEIEEE Access2169-35362023-01-0111172411725110.1109/ACCESS.2023.324492210043848A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT ImagesMohammad Rahil0https://orcid.org/0000-0002-5116-9127B. N. Anoop1https://orcid.org/0000-0002-6082-391XG. N. Girish2https://orcid.org/0000-0003-2101-2388Abhishek R. Kothari3https://orcid.org/0000-0003-0196-7021Shashidhar G. Koolagudi4Jeny Rajan5https://orcid.org/0000-0001-8045-6005Department of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, IndiaDepartment of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, IndiaDepartment of Computer Science and Engineering, Indian Institute of Information Technology Sri City, Chittoor, IndiaPink City Eye and Retina Center, Jaipur, IndiaDepartment of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, IndiaDepartment of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, IndiaRetinal Fluids (fluid collections) develop because of the accumulation of fluid in the retina, which may be caused by several retinal disorders, and can lead to loss of vision. Optical coherence tomography (OCT) provides non-invasive cross-sectional images of the retina and enables the visualization of different retinal abnormalities. The identification and segmentation of retinal cysts from OCT scans is gaining immense attention since the manual analysis of OCT data is time consuming and requires an experienced ophthalmologist. Identification and categorization of the retinal cysts aids in establishing the pathophysiology of various retinal diseases, such as macular edema, diabetic macular edema, and age-related macular degeneration. Hence, an automatic algorithm for the segmentation and detection of retinal cysts would be of great value to the ophthalmologists. In this study, we have proposed a convolutional neural network-based deep ensemble architecture that can segment the three different types of retinal cysts from the retinal OCT images. The quantitative and qualitative performance of the model was evaluated using the publicly available RETOUCH challenge dataset. The proposed model outperformed the state-of-the-art methods, with an overall improvement of 1.8%.https://ieeexplore.ieee.org/document/10043848/Optical coherence tomographyretinal cystsintra retinal fluidsub retinal fluidpigment epithelial detachmentensemble-approach |
| spellingShingle | Mohammad Rahil B. N. Anoop G. N. Girish Abhishek R. Kothari Shashidhar G. Koolagudi Jeny Rajan A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images IEEE Access Optical coherence tomography retinal cysts intra retinal fluid sub retinal fluid pigment epithelial detachment ensemble-approach |
| title | A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images |
| title_full | A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images |
| title_fullStr | A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images |
| title_full_unstemmed | A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images |
| title_short | A Deep Ensemble Learning-Based CNN Architecture for Multiclass Retinal Fluid Segmentation in OCT Images |
| title_sort | deep ensemble learning based cnn architecture for multiclass retinal fluid segmentation in oct images |
| topic | Optical coherence tomography retinal cysts intra retinal fluid sub retinal fluid pigment epithelial detachment ensemble-approach |
| url | https://ieeexplore.ieee.org/document/10043848/ |
| work_keys_str_mv | AT mohammadrahil adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT bnanoop adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT gngirish adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT abhishekrkothari adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT shashidhargkoolagudi adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT jenyrajan adeepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT mohammadrahil deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT bnanoop deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT gngirish deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT abhishekrkothari deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT shashidhargkoolagudi deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages AT jenyrajan deepensemblelearningbasedcnnarchitectureformulticlassretinalfluidsegmentationinoctimages |