An impact on convolutional neural networks amelioration for early detection of skin cancer | ||
| International Journal of Intelligent Computing and Information Sciences | ||
| Volume 24, Issue 2, June 2024, Pages 1-17 PDF (664.6 K) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/ijicis.2024.273450.1325 | ||
| Authors | ||
| Hany Maher Lala* 1; Hoda Saleh Ahmed2 | ||
| 1Department of Mathematics, Faculty of science, Al-Azhar University, Cairo, Egypt | ||
| 2Department of Mathematics, Faculty of science, Al-Azhar University (girls), Cairo, Egypt | ||
| Abstract | ||
| Skin cancer, one of the deadliest cancers globally, poses a significant threat to life. It is a type of tumor that originates in the skin and can spread to other areas of the body. Early detection significantly reduces mortality rates. Unfortunately, current diagnosis methods, primarily relying on visual inspection, lack accuracy. Deep learning techniques emerge as promising tools to aid dermatologists in achieving early and accurate skin cancer diagnosis. Specifically, convolutional neural networks (CNN) have emerged as the go-to method for tackling such challenges. The effectiveness of deep learning in medical image segmentation and detection surpasses the accuracy achieved by humans. By examining the latest research papers on the categorization of skin cancer through the utilization of convolutional neural networks, this comparative study delved into the subject matter. A comprehensive summary was given on the prevalent deep-learning models and datasets employed in the classification of skin cancer showing that convolutional neural networks may become a powerful tool for early detection of skin cancer and saving lives. | ||
| Keywords | ||
| Skin Lesion Detection; Deep learning; Segmentation; Classification | ||
|
Statistics Article View: 131 PDF Download: 161 |
||
