Comparative Analysis of Various Deep Learning Algorithms for Skin Cancer Disease

EasyChair Preprint 11697

Abstract

These days, skin illness among people has been a common illness, particularly in millions of individuals are enduring with different sorts of skin based illness. As a rule, these maladies have covered up perils which lead to human not as it were need of self-confidence and mental discouragement but too a chance of skin cancer. Determination of these sorts of illnesses as a rule required therapeutic specialists with high-level disobedient due to a need of visual determination in skin infection pictures. Additionally, manual determination of skin malady is frequently subjective, time-consuming, and required more human exertion. Skin cancer is one of the most hazardous types of cancer impacting millions of lives on a daily basis. Skin cancer originates by uncorrected (DNA) inside skin cells causing genetic mutation in the skin. Skin cancer spreads slowly to body parts and is therefore easier to treat in its early stages; so it's best to catch it in its early stages. The increase in skin cancer, high mortality rates and high medical costs require early detection of its symptoms.

Given the seriousness of these problems, scientists have developed many types of cancer at an early stage. Symmetry, colour, size, shape, etc. lesion parameters. It is used to identify skin cancers and distinguish between benign tumour’s and melanoma. This article provides detailed information about deep learning techniques for early skin detection. Research articles published in reputable journals on the subject of cancer screening were examined. Research findings are presented in the form of tools, graphs, tables, methods and procedures for better understanding

Keyphrases: DeepLearning, HAM-1000, KNN

@booklet{EasyChair:11697,
  author    = {Rohan Kukreti and Shivam Chawla and Sarthak Rawat and Sakshaat Dhiman and Gunjan Chhabra},
  title     = {Comparative Analysis of Various Deep Learning Algorithms for Skin Cancer Disease},
  howpublished = {EasyChair Preprint 11697},
  year      = {EasyChair, 2024}}