Binary Classification of Skin Cancer using Pretrained Deep Neural Networks | ||||
| Suez Canal Engineering, Energy and Environmental Science | ||||
| Volume 1, Issue 2, July 2023, Page 10-14 PDF (595.88 K) | ||||
| Document Type: Original Article | ||||
| DOI: 10.21608/sceee.2024.263585.1015 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Hadeer Hussein  1; Ahmed Magdy1; Rehab F. Abdel-Kader 2; Khaled Abd Elsalam Ali 1
| ||||
| 1Electrical Engineering Department, Faculty of Engineering, Suez Canal University, Ismailia, Egypt. | ||||
| 2Electrical Engineering Department, Faculty of Engineering, Port Said University, Port Said, Egypt. | ||||
| Abstract | ||||
| One of the most frequent kinds of cancer in the world is skin cancer. Clinical examination of skin lesions is essential to detect disease characteristics, but it is limited by long timeframes and a broad variety of interpretations. Computer vision is being used to detect diseases, help in diagnosis, and identify patient risks. This is particularly true for skin cancer, which may be lethal if not detected early on. Several computer-aided diagnosis and detection systems have already been developed to do this. Deep learning techniques have been developed to address these issues and assist dermatologists, as early and precise detection of skin cancer is critical to improve patient survival rates. In this paper, some pretrained deep neural networks are utilized for binary classification of skin cancer disease. They are used to classify between benign and malignant cancers in dermoscopic images. AlexNet, ResNet-18, SqueezeNet, and ShuffleNet are the used networks as transfer learning classifiers. In this study, we employed a Kaggle dataset titled "Skin Cancer: Malignant vs. Benign". The networks’ maximum accuracy approaches 89%. | ||||
| Keywords | ||||
| Binary classification; Deep learning; Pretrained deep neural networks; skin cancer | ||||
|
Statistics Article View: 116 PDF Download: 207 |
||||
