Open AccessClassification of Malignant Melanoma using Convolutional Neural Networks
Author(s) -
V. Vijeya Kaveri,
Vepa Meenakshi,
Kannan Karthik,
S Shri Raam
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1916/1/012070
Subject(s) - melanoma , skin cancer , convolutional neural network , segmentation , artificial intelligence , human skin , feature extraction , computer science , cancer , pattern recognition (psychology) , medicine , pathology , dermatology , biology , cancer research , genetics
Human cancer is one of the world’s most deadly diseases caused due to genetic disruption of skin cells and several molecular mutations. Skin cancer remains the most predominant form of cancer in human beings. The major goal is to detect skin cancer in early stages by research and analyse it using various techniques such as segmentation and feature extraction. The diagnosis of malignant melanoma skin cancer is done by dermatologist by examining skin and physical biopsy to determine the accurate stage of melanoma. It is developed because of high accumulation of melanin in the dermis layer of the skin. ABCD law is used in along with dermoscopy technology to detect malignant melanoma skin cancer. Image Acquisition Technique, pre-processing, segmentation, distinguishing function for skin Feature Selection, which specifies lesion characterization and classification methods are all conducted in this project for melanoma skin lesion characterization. We used symmetry detection, border detection, colour and diameter detection, as well as feature extraction to remove texture based features using a digital image processing technique. The deep Neural Network was proposed here to characterize the benign or malignant stage.