Detection of Melanoma Skin Cancer in Early Stages Using CNN Classification

EasyChair Preprint 9065

Abstract

Melanoma or Begins or skin cancer is one of the most dangerous type of skin disease. Skin exposure to UV radiation causes alterations or transformations in the melanocytes, which results in uncontrolled cell growth and causes burning or tanning of the skin. Even while some skin diseases are less common than others, they are dangerous because, if left untreated, they will likely progress and spread. In early stage it may not be treated properly.  The usual method of detecting skin is called a biopsy, which involves scraping off the patient's skin lesion and sending it for laboratory analysis. This approach is painful, challenging, and tiresome. Skin cancer from melanoma requires a dermatologist to devote a lot of effort to treating it. Here, in this study a new method is used which involves machine learning to identify skin cancer. With dermoscopy images of melanoma cancer, detecting of skin cancer is easily done with less human effort.

Keyphrases: Classification, Segmentation, feature extraction, image processing

@booklet{EasyChair:9065,
  author    = {Rupali Awhad and Dinesh Barode and Seema Kawathekar},
  title     = {Detection of Melanoma Skin Cancer in Early Stages Using CNN Classification},
  howpublished = {EasyChair Preprint 9065},
  year      = {EasyChair, 2022}}