Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification
Melanoma is widely recognized as one of the most lethal forms of skin cancer, with its incidence showing an upward trend in recent years. Nonetheless, the timely detection of this malignancy substantially enhances the likelihood of patients’ long-term survival. Several computer-based methods have re...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-09-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/13/17/2848 |
| _version_ | 1797582649277022208 |
|---|---|
| author | Tallha Akram Riaz Junejo Anas Alsuhaibani Muhammad Rafiullah Adeel Akram Nouf Abdullah Almujally |
| author_facet | Tallha Akram Riaz Junejo Anas Alsuhaibani Muhammad Rafiullah Adeel Akram Nouf Abdullah Almujally |
| author_sort | Tallha Akram |
| collection | DOAJ |
| description | Melanoma is widely recognized as one of the most lethal forms of skin cancer, with its incidence showing an upward trend in recent years. Nonetheless, the timely detection of this malignancy substantially enhances the likelihood of patients’ long-term survival. Several computer-based methods have recently been proposed, in the pursuit of diagnosing skin lesions at their early stages. Despite achieving some level of success, there still remains a margin of error that the machine learning community considers to be an unresolved research challenge. The primary objective of this study was to maximize the input feature information by combining multiple deep models in the first phase, and then to avoid noisy and redundant information by downsampling the feature set, using a novel evolutionary feature selection technique, in the second phase. By maintaining the integrity of the original feature space, the proposed idea generated highly discriminant feature information. Recent deep models, including Darknet53, DenseNet201, InceptionV3, and InceptionResNetV2, were employed in our study, for the purpose of feature extraction. Additionally, transfer learning was leveraged, to enhance the performance of our approach. In the subsequent phase, the extracted feature information from the chosen pre-existing models was combined, with the aim of preserving maximum information, prior to undergoing the process of feature selection, using a novel entropy-controlled gray wolf optimization (ECGWO) algorithm. The integration of fusion and selection techniques was employed, initially to incorporate the feature vector with a high level of information and, subsequently, to eliminate redundant and irrelevant feature information. The effectiveness of our concept is supported by an assessment conducted on three benchmark dermoscopic datasets: PH2, ISIC-MSK, and ISIC-UDA. In order to validate the proposed methodology, a comprehensive evaluation was conducted, including a rigorous comparison to established techniques in the field. |
| first_indexed | 2024-03-10T23:25:29Z |
| format | Article |
| id | doaj.art-255171b098aa497db527f343ea925ff8 |
| institution | Directory Open Access Journal |
| issn | 2075-4418 |
| language | English |
| last_indexed | 2024-03-10T23:25:29Z |
| publishDate | 2023-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Diagnostics |
| spelling | doaj.art-255171b098aa497db527f343ea925ff82023-11-19T08:00:24ZengMDPI AGDiagnostics2075-44182023-09-011317284810.3390/diagnostics13172848Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion ClassificationTallha Akram0Riaz Junejo1Anas Alsuhaibani2Muhammad Rafiullah3Adeel Akram4Nouf Abdullah Almujally5Department of Electrical and Computer Engineering, COMSATS University Islamabad, Wah Cantt Campus, Islamabad 45040, PakistanDepartment of Electrical and Computer Engineering, COMSATS University Islamabad, Wah Cantt Campus, Islamabad 45040, PakistanDepartment of Information Systems, College of Computer Engineering and Sciences, Prince Sattam bin Abdulaziz University, Al-Kharj 11942, Saudi ArabiaDepartment of Mathematics, COMSATS University Islamabad, Lahore Campus, Lahore 54000, PakistanDepartment of Electrical and Computer Engineering, COMSATS University Islamabad, Wah Cantt Campus, Islamabad 45040, PakistanDepartment of Information Systems, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University, Riyadh 11671, Saudi ArabiaMelanoma is widely recognized as one of the most lethal forms of skin cancer, with its incidence showing an upward trend in recent years. Nonetheless, the timely detection of this malignancy substantially enhances the likelihood of patients’ long-term survival. Several computer-based methods have recently been proposed, in the pursuit of diagnosing skin lesions at their early stages. Despite achieving some level of success, there still remains a margin of error that the machine learning community considers to be an unresolved research challenge. The primary objective of this study was to maximize the input feature information by combining multiple deep models in the first phase, and then to avoid noisy and redundant information by downsampling the feature set, using a novel evolutionary feature selection technique, in the second phase. By maintaining the integrity of the original feature space, the proposed idea generated highly discriminant feature information. Recent deep models, including Darknet53, DenseNet201, InceptionV3, and InceptionResNetV2, were employed in our study, for the purpose of feature extraction. Additionally, transfer learning was leveraged, to enhance the performance of our approach. In the subsequent phase, the extracted feature information from the chosen pre-existing models was combined, with the aim of preserving maximum information, prior to undergoing the process of feature selection, using a novel entropy-controlled gray wolf optimization (ECGWO) algorithm. The integration of fusion and selection techniques was employed, initially to incorporate the feature vector with a high level of information and, subsequently, to eliminate redundant and irrelevant feature information. The effectiveness of our concept is supported by an assessment conducted on three benchmark dermoscopic datasets: PH2, ISIC-MSK, and ISIC-UDA. In order to validate the proposed methodology, a comprehensive evaluation was conducted, including a rigorous comparison to established techniques in the field.https://www.mdpi.com/2075-4418/13/17/2848convolutional neural networksfeature selectiontransfer learningfeature fusiongray wolf optimizationdeep learning |
| spellingShingle | Tallha Akram Riaz Junejo Anas Alsuhaibani Muhammad Rafiullah Adeel Akram Nouf Abdullah Almujally Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification Diagnostics convolutional neural networks feature selection transfer learning feature fusion gray wolf optimization deep learning |
| title | Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification |
| title_full | Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification |
| title_fullStr | Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification |
| title_full_unstemmed | Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification |
| title_short | Precision in Dermatology: Developing an Optimal Feature Selection Framework for Skin Lesion Classification |
| title_sort | precision in dermatology developing an optimal feature selection framework for skin lesion classification |
| topic | convolutional neural networks feature selection transfer learning feature fusion gray wolf optimization deep learning |
| url | https://www.mdpi.com/2075-4418/13/17/2848 |
| work_keys_str_mv | AT tallhaakram precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification AT riazjunejo precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification AT anasalsuhaibani precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification AT muhammadrafiullah precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification AT adeelakram precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification AT noufabdullahalmujally precisionindermatologydevelopinganoptimalfeatureselectionframeworkforskinlesionclassification |