MELANOMA SKIN CANCER DETECTION USING DEEP LEARNING AND CLASSICAL MACHINE LEARNING TECHNIQUES: A HYBRID APPROACH

MELANOMA SKIN CANCER DETECTION USING DEEP LEARNING AND CLASSICAL MACHINE LEARNING TECHNIQUES: A HYBRID APPROACH

Krupa H Malladad*, Mr. Arun Raj S R **

*   MTech Research Scholar, Department of Electronics and Communication Engineering

** Assistant Professor, Department of Electronics and Communication Engineering

University BDT College of Engineering, Davangere 577004, Karnataka, India

Abstract- There are various types of cancers, with skin cancer being the most common. In general, the number of people infected with skin cancer each year is less than the number of people infected with all other types of cancer combined. The global mortality rate for pores and skin cancer has risen. According to the arena health company, early detection of skin variations greatly improves the chances of exact treatment and treatment so that the case can be saved. The laptop computer is integrated with deep learning software, specifically convolutional neural networks (CNN), we dragged this Deep learning armature to expand a model that categorizes the given inflamed pores and skin image of case as nasty (carcinoma or dangerous) or benign (inoffensive) by way of utilizing various Python libraries. This interpretation is taught and tested using data from the Global Skin Imaging Collaboration (ISIC). The primary goal of this model is to detect skin cancer in early stages and treat it effectively in order to reduce mortality.

Index Terms- Detection of melanoma Deep understanding, Classical machine mastery, and CNN