EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques
Abstract Semantic segmentation involves an imminent part in the investigation of medical images, particularly in the domain of microvascular decompression, where publicly available datasets are scarce, and expert annotation is demanding. In response to this challenge, this study presents a meticulou...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-05-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-02470-5 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850125854391140352 |
|---|---|
| author | B. Dhiyanesh M. Vijayalakshmi P. Saranya D. Viji |
| author_facet | B. Dhiyanesh M. Vijayalakshmi P. Saranya D. Viji |
| author_sort | B. Dhiyanesh |
| collection | DOAJ |
| description | Abstract Semantic segmentation involves an imminent part in the investigation of medical images, particularly in the domain of microvascular decompression, where publicly available datasets are scarce, and expert annotation is demanding. In response to this challenge, this study presents a meticulously curated dataset comprising 2003 RGB microvascular decompression images, each intricately paired with annotated masks. Extensive data preprocessing and augmentation strategies were employed to fortify the training dataset, enhancing the robustness of proposed deep learning model. Numerous up-to-date semantic segmentation approaches, including DeepLabv3+, U-Net, DilatedFastFCN with JPU, DANet, and a custom Vanilla architecture, were trained and evaluated using diverse performance metrics. Among these models, DeepLabv3 + emerged as a strong contender, notably excelling in F1 score. Innovatively, ensemble techniques, such as stacking and bagging, were introduced to further elevate segmentation performance. Bagging, notably with the Naïve Bayes approach, exhibited significant improvements, underscoring the potential of ensemble methods in medical image segmentation. The proposed EnsembleEdgeFusion technique exhibited superior loss reduction during training compared to DeepLabv3 + and achieved maximum Mean Intersection over Union (MIoU) scores of 77.73%, surpassing other models. Category-wise analysis affirmed its superiority in accurately delineating various categories within the test dataset. |
| format | Article |
| id | doaj-art-d65eace26b184843958a89e215fd15bb |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-d65eace26b184843958a89e215fd15bb2025-08-20T02:34:02ZengNature PortfolioScientific Reports2045-23222025-05-0115112210.1038/s41598-025-02470-5EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniquesB. Dhiyanesh0M. Vijayalakshmi1P. Saranya2D. Viji3Department of Computer Science and Engineering (ETech), SRM Institute of Science and Technology, Vadapalani CampusDepartment of Computing Technologies, SRM Institute of Science and Technology, KattankulathurDepartment of Computational Intelligence, SRM Institute of Science and Technology, KattankulathurDepartment of Computing Technologies, SRM Institute of Science and Technology, KattankulathurAbstract Semantic segmentation involves an imminent part in the investigation of medical images, particularly in the domain of microvascular decompression, where publicly available datasets are scarce, and expert annotation is demanding. In response to this challenge, this study presents a meticulously curated dataset comprising 2003 RGB microvascular decompression images, each intricately paired with annotated masks. Extensive data preprocessing and augmentation strategies were employed to fortify the training dataset, enhancing the robustness of proposed deep learning model. Numerous up-to-date semantic segmentation approaches, including DeepLabv3+, U-Net, DilatedFastFCN with JPU, DANet, and a custom Vanilla architecture, were trained and evaluated using diverse performance metrics. Among these models, DeepLabv3 + emerged as a strong contender, notably excelling in F1 score. Innovatively, ensemble techniques, such as stacking and bagging, were introduced to further elevate segmentation performance. Bagging, notably with the Naïve Bayes approach, exhibited significant improvements, underscoring the potential of ensemble methods in medical image segmentation. The proposed EnsembleEdgeFusion technique exhibited superior loss reduction during training compared to DeepLabv3 + and achieved maximum Mean Intersection over Union (MIoU) scores of 77.73%, surpassing other models. Category-wise analysis affirmed its superiority in accurately delineating various categories within the test dataset.https://doi.org/10.1038/s41598-025-02470-5Semantic segmentationMicrovascular decompressionDeep learningEnsemble techniquesMedical image analysisBagging |
| spellingShingle | B. Dhiyanesh M. Vijayalakshmi P. Saranya D. Viji EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques Scientific Reports Semantic segmentation Microvascular decompression Deep learning Ensemble techniques Medical image analysis Bagging |
| title | EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| title_full | EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| title_fullStr | EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| title_full_unstemmed | EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| title_short | EnsembleEdgeFusion: advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| title_sort | ensembleedgefusion advancing semantic segmentation in microvascular decompression imaging with innovative ensemble techniques |
| topic | Semantic segmentation Microvascular decompression Deep learning Ensemble techniques Medical image analysis Bagging |
| url | https://doi.org/10.1038/s41598-025-02470-5 |
| work_keys_str_mv | AT bdhiyanesh ensembleedgefusionadvancingsemanticsegmentationinmicrovasculardecompressionimagingwithinnovativeensembletechniques AT mvijayalakshmi ensembleedgefusionadvancingsemanticsegmentationinmicrovasculardecompressionimagingwithinnovativeensembletechniques AT psaranya ensembleedgefusionadvancingsemanticsegmentationinmicrovasculardecompressionimagingwithinnovativeensembletechniques AT dviji ensembleedgefusionadvancingsemanticsegmentationinmicrovasculardecompressionimagingwithinnovativeensembletechniques |