Title: Comparative approach for discovery of cancerous skin using deep structured learning

Authors: K.A. Varun Kumar; Sree T. Sucharitha; R. Priyadarshini; N. Rajendran

Addresses: Department of Networking and Communications, School of Computing, SRM Institute of Science and Technology, Kattankulathur, Chennai, 603203, India ' Department of Community Medicine, Annaii Medical College and Hospital, Chennai, 600107, Tamil Nadu, India ' School of Computer Science and Engineering, Vellore institute of Technology, Chennai, 600127, India ' Department of Information Technology, B S Abdur Rahman Crescent Institute of Science and Technology, 600048, India

Abstract: In recent days, skin cancer cases are significantly increasing due to ozone layer damage. Impacts of the ozone layer damage ultraviolet rays directly penetrate the human skin leading to skin cancer. For the above reason it is important to develop the new model to detect the skin cancer in early stage from the digital data and image processing techniques. The research in detection of skin cancer is highly active from the year 2016. In this paper we attempt both the machine learning and deep learning algorithm to detect the skin cancer to improve the detection accuracy and early diagnosis of patients. In this proposed model we use machine learning algorithms like Naive Bayes, decision tree and KNN in that decision tree algorithm outperforms the rest of the algorithms used with accuracy of 83%. To further improve the accuracy we proposed the deep learning approach such as convolutional neural network to automate the skin cancer detection. In this paper we also compare the model accuracy of 93.54%.

Keywords: skin cancer; convolutional neural network; Naives Bayes; decision tree; KNN; cancer detection; deep learning.

DOI: 10.1504/IJNT.2023.134030

International Journal of Nanotechnology, 2023 Vol.20 No.5/6/7/8/9/10, pp.744 - 758

Received: 17 Nov 2021
Accepted: 22 Feb 2022

Published online: 10 Oct 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article