|
|
Bimonthly Since 1986 |
ISSN 1004-9037
|
|
|
|
|
Publication Details |
Edited by: Editorial Board of Journal of Data Acquisition and Processing
P.O. Box 2704, Beijing 100190, P.R. China
Sponsored by: Institute of Computing Technology, CAS & China Computer Federation
Undertaken by: Institute of Computing Technology, CAS
Published by: SCIENCE PRESS, BEIJING, CHINA
Distributed by:
China: All Local Post Offices
|
|
|
|
|
|
|
|
|
|
Abstract
Skin cancer is common and dangerous. This illness, like other cancers, requires early detection. Conventional skin cancer diagnosis methods are inaccurate and can lead to unnecessary inspections. Certain cancer detection machine learning algorithms support a limited number of skin cancer classifications, which might be a drawback. The research method can automatically detect skin cancer and benign tumor lesions using the Convolutional Neural Network. The model contains three hidden layers with 16–32–64 output channels. The model uses SGD, RMSprop, Adam, and Nadam optimizers with a learning rate of 0.001. The Adam optimizer classifies ISIC dataset skin lesions as benign or malignant with 93% accuracy. The results outperform the current skin cancer classification approach. This study uses ISIC [24].
Keyword
Skin Cancer , ISIC , Convolutional Neural Network, Adam, and Nadam.
PDF Download (click here)
|
|
|
|
|