SkCanNet: A Deep Learning based Skin Cancer Classification Approach
Document Type
Article
Publication Title
Annals of Emerging Technologies in Computing
Abstract
Skin Cancer classification has been one of the most challenging problems for dermatologists; it is a tremendously tedious process to detect the kind of lesion/cancer form it is for just the human eye. Deep learning has become popular due to its potential to learn complex traits from the huge dataset. A prominent deep learning model for image categorization is the convolutional neural network (CNN). Many researchers have been conducted on the efficiency of CNN’s use to classify skin cancer forms. In this paper, the efficiency of VGG bottleneck features and transfer learning have been used on 3 kinds of skin cancers namely, (a) squamous cell carcinoma, (b) basal cell carcinoma and (c) melanoma. The proposed model comprises of VGG-16 NET and Transfer Learning with 2 fully-connected layers. The proposed model is experimented on 1077 dermoscopy images in total (MSK-1, UDA-1, UDA-2, HAM10000). The experimental analysis proves that the proposed model achieves higher values for accuracy, specificity and sensitivity.
First Page
35
Last Page
45
DOI
10.33166/AETiC.2023.04.004
Publication Date
10-1-2023
Recommended Citation
Onesimu, J. Andrew; Nair, Varun Unnikrishnan; Sagayam, Martin K.; and Eunice, Jennifer, "SkCanNet: A Deep Learning based Skin Cancer Classification Approach" (2023). Open Access archive. 7769.
https://impressions.manipal.edu/open-access-archive/7769