A novel approach toward skin cancer classification through fused deep features and neutrosophic environment
Variations in the size and texture of melanoma make the classification procedure more complex in a computer-aided diagnostic (CAD) system. The research proposes an innovative hybrid deep learning-based layer-fusion and neutrosophic-set technique for identifying skin lesions. The off-the-shelf networ...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2023-04-01
|
| Series: | Frontiers in Public Health |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fpubh.2023.1123581/full |
| _version_ | 1797845352073658368 |
|---|---|
| author | Ahmed Abdelhafeez Hoda K. Mohamed Ali Maher Nariman A. Khalil |
| author_facet | Ahmed Abdelhafeez Hoda K. Mohamed Ali Maher Nariman A. Khalil |
| author_sort | Ahmed Abdelhafeez |
| collection | DOAJ |
| description | Variations in the size and texture of melanoma make the classification procedure more complex in a computer-aided diagnostic (CAD) system. The research proposes an innovative hybrid deep learning-based layer-fusion and neutrosophic-set technique for identifying skin lesions. The off-the-shelf networks are examined to categorize eight types of skin lesions using transfer learning on International Skin Imaging Collaboration (ISIC) 2019 skin lesion datasets. The top two networks, which are GoogleNet and DarkNet, achieved an accuracy of 77.41 and 82.42%, respectively. The proposed method works in two successive stages: first, boosting the classification accuracy of the trained networks individually. A suggested feature fusion methodology is applied to enrich the extracted features’ descriptive power, which promotes the accuracy to 79.2 and 84.5%, respectively. The second stage explores how to combine these networks for further improvement. The error-correcting output codes (ECOC) paradigm is utilized for constructing a set of well-trained true and false support vector machine (SVM) classifiers via fused DarkNet and GoogleNet feature maps, respectively. The ECOC’s coding matrices are designed to train each true classifier and its opponent in a one-versus-other fashion. Consequently, contradictions between true and false classifiers in terms of their classification scores create an ambiguity zone quantified by the indeterminacy set. Recent neutrosophic techniques resolve this ambiguity to tilt the balance toward the correct skin cancer class. As a result, the classification score is increased to 85.74%, outperforming the recent proposals by an obvious step. The trained models alongside the implementation of the proposed single-valued neutrosophic sets (SVNSs) will be publicly available for aiding relevant research fields. |
| first_indexed | 2024-04-09T17:37:39Z |
| format | Article |
| id | doaj.art-b08d91071f14401fa553012e1dde5c2a |
| institution | Directory Open Access Journal |
| issn | 2296-2565 |
| language | English |
| last_indexed | 2024-04-09T17:37:39Z |
| publishDate | 2023-04-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Public Health |
| spelling | doaj.art-b08d91071f14401fa553012e1dde5c2a2023-04-17T13:36:25ZengFrontiers Media S.A.Frontiers in Public Health2296-25652023-04-011110.3389/fpubh.2023.11235811123581A novel approach toward skin cancer classification through fused deep features and neutrosophic environmentAhmed Abdelhafeez0Hoda K. Mohamed1Ali Maher2Nariman A. Khalil3Faculty of Information Systems and Computer Science, October 6th University, Cairo, EgyptFaculty of Engineering, Ain Shams University, Cairo, EgyptMilitary Technical College, Cairo, EgyptFaculty of Information Systems and Computer Science, October 6th University, Cairo, EgyptVariations in the size and texture of melanoma make the classification procedure more complex in a computer-aided diagnostic (CAD) system. The research proposes an innovative hybrid deep learning-based layer-fusion and neutrosophic-set technique for identifying skin lesions. The off-the-shelf networks are examined to categorize eight types of skin lesions using transfer learning on International Skin Imaging Collaboration (ISIC) 2019 skin lesion datasets. The top two networks, which are GoogleNet and DarkNet, achieved an accuracy of 77.41 and 82.42%, respectively. The proposed method works in two successive stages: first, boosting the classification accuracy of the trained networks individually. A suggested feature fusion methodology is applied to enrich the extracted features’ descriptive power, which promotes the accuracy to 79.2 and 84.5%, respectively. The second stage explores how to combine these networks for further improvement. The error-correcting output codes (ECOC) paradigm is utilized for constructing a set of well-trained true and false support vector machine (SVM) classifiers via fused DarkNet and GoogleNet feature maps, respectively. The ECOC’s coding matrices are designed to train each true classifier and its opponent in a one-versus-other fashion. Consequently, contradictions between true and false classifiers in terms of their classification scores create an ambiguity zone quantified by the indeterminacy set. Recent neutrosophic techniques resolve this ambiguity to tilt the balance toward the correct skin cancer class. As a result, the classification score is increased to 85.74%, outperforming the recent proposals by an obvious step. The trained models alongside the implementation of the proposed single-valued neutrosophic sets (SVNSs) will be publicly available for aiding relevant research fields.https://www.frontiersin.org/articles/10.3389/fpubh.2023.1123581/fulltransfer learningfused deep featuresskin cancer classificationmulti-support vector machineerror-correcting output codessingle-valued neutrosophic sets |
| spellingShingle | Ahmed Abdelhafeez Hoda K. Mohamed Ali Maher Nariman A. Khalil A novel approach toward skin cancer classification through fused deep features and neutrosophic environment Frontiers in Public Health transfer learning fused deep features skin cancer classification multi-support vector machine error-correcting output codes single-valued neutrosophic sets |
| title | A novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| title_full | A novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| title_fullStr | A novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| title_full_unstemmed | A novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| title_short | A novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| title_sort | novel approach toward skin cancer classification through fused deep features and neutrosophic environment |
| topic | transfer learning fused deep features skin cancer classification multi-support vector machine error-correcting output codes single-valued neutrosophic sets |
| url | https://www.frontiersin.org/articles/10.3389/fpubh.2023.1123581/full |
| work_keys_str_mv | AT ahmedabdelhafeez anovelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT hodakmohamed anovelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT alimaher anovelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT narimanakhalil anovelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT ahmedabdelhafeez novelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT hodakmohamed novelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT alimaher novelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment AT narimanakhalil novelapproachtowardskincancerclassificationthroughfuseddeepfeaturesandneutrosophicenvironment |