A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures
<b>Background:</b> Retinal blood vessel segmentation plays an important role in diagnosing retinal diseases such as diabetic retinopathy, glaucoma, and hypertensive retinopathy. Accurate segmentation of blood vessels in retinal images presents a challenging task due to noise, low contras...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-01-01
|
| Series: | Biomedicines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-9059/13/1/141 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588967089799168 |
|---|---|
| author | Mohamed Chetoui Moulay A. Akhloufi |
| author_facet | Mohamed Chetoui Moulay A. Akhloufi |
| author_sort | Mohamed Chetoui |
| collection | DOAJ |
| description | <b>Background:</b> Retinal blood vessel segmentation plays an important role in diagnosing retinal diseases such as diabetic retinopathy, glaucoma, and hypertensive retinopathy. Accurate segmentation of blood vessels in retinal images presents a challenging task due to noise, low contrast, and the complex morphology of blood vessel structures. <b>Methods:</b> In this study, we propose a novel ensemble learning framework combining four deep learning architectures: U-Net, ResNet50, U-Net with a ResNet50 backbone, and U-Net with a transformer block. Each architecture is customized to enhance feature extraction and segmentation performance. The models are trained on the DRIVE and STARE datasets to improve the degree of generalization and evaluated using the performance metric accuracy, F1-Score, sensitivity, specificity, and AUC. <b>Results:</b> The ensemble meta-model integrates predictions from these architectures using a stacking approach, achieving state-of-the-art performance with an accuracy of 0.9778, an AUC of 0.9912, and an F1-Score of 0.8231. These results demonstrate the performance of the proposed technique in identifying thin retinal blood vessels. <b>Conclusions:</b> A comparative analysis using qualitative and quantitative results with individual models highlights the robustness of the ensemble framework, especially under conditions of noise and poor visibility. |
| format | Article |
| id | doaj-art-4835641ab163484689a383dbbed179c6 |
| institution | Kabale University |
| issn | 2227-9059 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biomedicines |
| spelling | doaj-art-4835641ab163484689a383dbbed179c62025-01-24T13:24:09ZengMDPI AGBiomedicines2227-90592025-01-0113114110.3390/biomedicines13010141A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning ArchitecturesMohamed Chetoui0Moulay A. Akhloufi1Perception, Robotics, and Intelligent Machines Lab (PRIME), Department of Computer Science, Université de Moncton, Moncton, NB E1A 3E9, CanadaPerception, Robotics, and Intelligent Machines Lab (PRIME), Department of Computer Science, Université de Moncton, Moncton, NB E1A 3E9, Canada<b>Background:</b> Retinal blood vessel segmentation plays an important role in diagnosing retinal diseases such as diabetic retinopathy, glaucoma, and hypertensive retinopathy. Accurate segmentation of blood vessels in retinal images presents a challenging task due to noise, low contrast, and the complex morphology of blood vessel structures. <b>Methods:</b> In this study, we propose a novel ensemble learning framework combining four deep learning architectures: U-Net, ResNet50, U-Net with a ResNet50 backbone, and U-Net with a transformer block. Each architecture is customized to enhance feature extraction and segmentation performance. The models are trained on the DRIVE and STARE datasets to improve the degree of generalization and evaluated using the performance metric accuracy, F1-Score, sensitivity, specificity, and AUC. <b>Results:</b> The ensemble meta-model integrates predictions from these architectures using a stacking approach, achieving state-of-the-art performance with an accuracy of 0.9778, an AUC of 0.9912, and an F1-Score of 0.8231. These results demonstrate the performance of the proposed technique in identifying thin retinal blood vessels. <b>Conclusions:</b> A comparative analysis using qualitative and quantitative results with individual models highlights the robustness of the ensemble framework, especially under conditions of noise and poor visibility.https://www.mdpi.com/2227-9059/13/1/141deep learningblood vesselU-Netdiabetic retinopathyimage segmentationmedical imaging |
| spellingShingle | Mohamed Chetoui Moulay A. Akhloufi A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures Biomedicines deep learning blood vessel U-Net diabetic retinopathy image segmentation medical imaging |
| title | A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures |
| title_full | A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures |
| title_fullStr | A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures |
| title_full_unstemmed | A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures |
| title_short | A Novel Ensemble Meta-Model for Enhanced Retinal Blood Vessel Segmentation Using Deep Learning Architectures |
| title_sort | novel ensemble meta model for enhanced retinal blood vessel segmentation using deep learning architectures |
| topic | deep learning blood vessel U-Net diabetic retinopathy image segmentation medical imaging |
| url | https://www.mdpi.com/2227-9059/13/1/141 |
| work_keys_str_mv | AT mohamedchetoui anovelensemblemetamodelforenhancedretinalbloodvesselsegmentationusingdeeplearningarchitectures AT moulayaakhloufi anovelensemblemetamodelforenhancedretinalbloodvesselsegmentationusingdeeplearningarchitectures AT mohamedchetoui novelensemblemetamodelforenhancedretinalbloodvesselsegmentationusingdeeplearningarchitectures AT moulayaakhloufi novelensemblemetamodelforenhancedretinalbloodvesselsegmentationusingdeeplearningarchitectures |