ISSN (Online) : 2456 - 0774

Email : ijasret@gmail.com

ISSN (Online) 2456 - 0774


DETECTION OF SKIN CANCER USING CNN ALGORITHM

Abstract

Abstract: Skin cancer is one of the deadliest types of cancer. If it is not diagnosed and treated early on, it is likely to spread to other areas of the body. It is primarily caused by abnormal skin cell development, which occurs often when the body is exposed to sunlight. The Surveillance Furthermore, identifying skin malignant development in its early stages is an expensive and difficult process. It is graded according to where it grows and what type of cell it is. The classification of lesions necessitates a high level of precision and recall. The MNIST HAM-10000 dataset containing dermoscopy images will be included in this article. The aim is to propose a method that uses a Convolution Neural Network to diagnose skin cancer and classify it into various groups. Image recognition and a deep learning algorithm are used in the diagnosis process. The noise and picture resolution were removed from the dermoscopy shot of skin cancer that was taken. Using different image augmentation methods, the image count may also be improved. Finally, the Transfer Learning approach is used to improve the image recognition accuracy even further. The weighted average Precision of our CNN model was 0.88, the weighted average Recall was 0.74, and the weighted f1-score was 0.77. The accuracy of the transfer learning method using the ResNet model was 90.51 percent.Keywords: Skin Cancer, Skin lesion, Deep learning, CNN.

Full Text PDF

IMPORTANT DATES 

Submit paper at ijasret@gmail.com

Paper Submission Open For October 2024
UGC indexed in (Old UGC) 2017
Last date for paper submission 30th October, 2024
Deadline Submit Paper any time
Publication of Paper Within 15-30 Days after completing all the formalities
Publication Fees  Rs.6000 (UG student)
Publication Fees  Rs.8000 (PG student)