Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis
This research paper presents a deep-learning approach to early detection of skin cancer using image augmentation techniques. We introduce a two-stage image augmentation process utilizing geometric augmentation and a generative adversarial network (GAN) to differentiate skin cancer categories. The pu...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-12-01
|
| Series: | Computation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-3197/11/12/246 |
| _version_ | 1797381522022465536 |
|---|---|
| author | Catur Supriyanto Abu Salam Junta Zeniarja Adi Wijaya |
| author_facet | Catur Supriyanto Abu Salam Junta Zeniarja Adi Wijaya |
| author_sort | Catur Supriyanto |
| collection | DOAJ |
| description | This research paper presents a deep-learning approach to early detection of skin cancer using image augmentation techniques. We introduce a two-stage image augmentation process utilizing geometric augmentation and a generative adversarial network (GAN) to differentiate skin cancer categories. The public HAM10000 dataset was used to test how well the proposed model worked. Various pre-trained convolutional neural network (CNN) models, including Xception, Inceptionv3, Resnet152v2, EfficientnetB7, InceptionresnetV2, and VGG19, were employed. Our approach demonstrates an accuracy of 96.90%, precision of 97.07%, recall of 96.87%, and F1-score of 96.97%, surpassing the performance of other state-of-the-art methods. The paper also discusses the use of Shapley Additive Explanations (SHAP), an interpretable technique for skin cancer diagnosis, which can help clinicians understand the reasoning behind the diagnosis and improve trust in the system. Overall, the proposed method presents a promising approach to automated skin cancer detection that could improve patient outcomes and reduce healthcare costs. |
| first_indexed | 2024-03-08T20:52:47Z |
| format | Article |
| id | doaj.art-7799f4692e134c16b931d4f19773590e |
| institution | Directory Open Access Journal |
| issn | 2079-3197 |
| language | English |
| last_indexed | 2024-03-08T20:52:47Z |
| publishDate | 2023-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computation |
| spelling | doaj.art-7799f4692e134c16b931d4f19773590e2023-12-22T14:01:20ZengMDPI AGComputation2079-31972023-12-01111224610.3390/computation11120246Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer DiagnosisCatur Supriyanto0Abu Salam1Junta Zeniarja2Adi Wijaya3Faculty of Computer Science, Universitas Dian Nuswantoro, Semarang 50131, IndonesiaFaculty of Computer Science, Universitas Dian Nuswantoro, Semarang 50131, IndonesiaFaculty of Computer Science, Universitas Dian Nuswantoro, Semarang 50131, IndonesiaDepartment of Health Information Management, Universitas Indonesia Maju, Jakarta 12610, IndonesiaThis research paper presents a deep-learning approach to early detection of skin cancer using image augmentation techniques. We introduce a two-stage image augmentation process utilizing geometric augmentation and a generative adversarial network (GAN) to differentiate skin cancer categories. The public HAM10000 dataset was used to test how well the proposed model worked. Various pre-trained convolutional neural network (CNN) models, including Xception, Inceptionv3, Resnet152v2, EfficientnetB7, InceptionresnetV2, and VGG19, were employed. Our approach demonstrates an accuracy of 96.90%, precision of 97.07%, recall of 96.87%, and F1-score of 96.97%, surpassing the performance of other state-of-the-art methods. The paper also discusses the use of Shapley Additive Explanations (SHAP), an interpretable technique for skin cancer diagnosis, which can help clinicians understand the reasoning behind the diagnosis and improve trust in the system. Overall, the proposed method presents a promising approach to automated skin cancer detection that could improve patient outcomes and reduce healthcare costs.https://www.mdpi.com/2079-3197/11/12/246deep learningskin cancerimage augmentationGANgeometric augmentationimage classification |
| spellingShingle | Catur Supriyanto Abu Salam Junta Zeniarja Adi Wijaya Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis Computation deep learning skin cancer image augmentation GAN geometric augmentation image classification |
| title | Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis |
| title_full | Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis |
| title_fullStr | Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis |
| title_full_unstemmed | Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis |
| title_short | Two-Stage Input-Space Image Augmentation and Interpretable Technique for Accurate and Explainable Skin Cancer Diagnosis |
| title_sort | two stage input space image augmentation and interpretable technique for accurate and explainable skin cancer diagnosis |
| topic | deep learning skin cancer image augmentation GAN geometric augmentation image classification |
| url | https://www.mdpi.com/2079-3197/11/12/246 |
| work_keys_str_mv | AT catursupriyanto twostageinputspaceimageaugmentationandinterpretabletechniqueforaccurateandexplainableskincancerdiagnosis AT abusalam twostageinputspaceimageaugmentationandinterpretabletechniqueforaccurateandexplainableskincancerdiagnosis AT juntazeniarja twostageinputspaceimageaugmentationandinterpretabletechniqueforaccurateandexplainableskincancerdiagnosis AT adiwijaya twostageinputspaceimageaugmentationandinterpretabletechniqueforaccurateandexplainableskincancerdiagnosis |