Melanoma Detection and Classification using Deep Learning

Melanoma Detection and Classification using Deep Learning

1 Bhavani C N, 2 Dr Shankaragowda B B

1Sem MCA Student, Department of MCA, BIET, Davanagere

2Associate Professor and HOD, Department of MCA, BIET, Davanagere

ABSTRACT: Melanoma is a type of carcinoma with a notably high mortality rate. Accurate diagnosis of this aggressive cancer is crucial due to its severe implications. Key diagnostic indicators include asymmetrical shape, heterogeneous color, diameter greater than 6 mm, and irregular borders, which dermatologists typically identify through visual examination. The conventional method for carcinoma detection is biopsy, involving the removal or scraping of skin samples for extensive laboratory testing. This process is both painful and time-consuming. To improve patient experience and enhance diagnostic efficiency, computer-based detection using image processing techniques and deep learning algorithms, specifically Convolutional Neural Networks (CNNs), has been developed to accurately identify melanoma.
Keywords: Deep learning, CNN, Computer-based detection