An Efficient Framework using Deep Learning for Skin Cancer Classification
Sarthak Garg1, Harshit Garg2, Vikas Tripathi3, Bhasker Pant4, Kumud Pant5
1Sarthak Garg, Department of Computer Science and Engineering, Graphic Era deemed to be University, Dehradun, India.
2Harshit Garg, Department of Computer Science and Engineering, Graphic Era deemed to be university, Dehradun, India.
3Vikas Tripathi, Department of Computer Science and Engineering, Graphic Era deemed to be university, Dehradun, India.
4Bhasker Pant, Department of Computer Science and Engineering, Graphic Era deemed to be university, Dehradun, India.
5Kumud Pant, Department of Computer Science and Engineering, Graphic Era deemed to be university, Dehradun, India.
Manuscript received on November 13, 2019. | Revised Manuscript received on 22 November, 2019. | Manuscript published on December 10, 2019. | PP: 1947-1951 | Volume-9 Issue-2, December 2019. | Retrieval Number: B7854129219/2019©BEIESP | DOI: 10.35940/ijitee.B7854.129219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: A good image analysis model can be very helpful in accurate diagnosis/classification of diseases for which images are available. Due to a plethora of public image databases, training and testing of algorithms on the dataset have helped in the development of an efficient framework for image classification. Skin cancer is one such disease for which recently image databases have been developed. Of the various methods of classification of skin cancer based on image analysis, convolutional neural network (CNN) has proven to be better performing than conventional machine learning approach. Realizing the importance of developing an efficient framework for skin cancer classification, this paper proposes a framework which utilizes VGG-16 CNN model to classify cancer images into categories namely, malignant or benign. We have trained the model using the skin cancer images freely accessible on the ISIC archive and attained an accuracy of 97.81%.
Keywords: VGG-16 Model, Convolutional Neural Network, Deep Learning, ReLu, skin cancer, image processing.
Scope of the Article: Deep Learning