• Prof. D. S. Zingade
• Aditya Gujar
• Swaraj Govindwar
• Sujeet Bhagwat

In recent years, skin cancer is seen as one of the most perilous form of cancers in human beings. Melanoma, a type of skin cancer, is capable of deep invasion. The most dangerous characteristic of it, is that it spreads widely over the body through the lymphatic vessels and blood vessels. Therefore, early detection and diagnosis of melanoma is an important factor for the prognosis of the disease. Recently many deep learning techniques are emerging which enable development of intelligent medical image-based diagnosis systems that can assist or help oncologists in making better decisions. This paper presents a model which will focus on skin lesion classification. Here, machine learning techniques, especially neural networks, will be used for early melanoma detection by classifying the image as malignant(cancerous) or benign(non-cancerous). The process of detection will include image processing methods like image conversion and feature extraction. The processed image will then be used to train a ML model which will be developed using Artificial Neural Networks (ANN) like Back Propagation and Feed Forward networks. The trained model can be used for classification of an image as cancerous or non-cancerous.

Melanoma, Image Processing, Artificial Neural Networks, Feed Forward, Back Propagation, Machine Learning, Dermoscopy, Skin Lesion, Classification

"PREDICTION OF MELENOMA CANCER USING NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.139-142, April-2019, Available :http://www.jetir.org/papers/JETIR1904P24.pdf

"PREDICTION OF MELENOMA CANCER USING NEURAL NETWORKS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp139-142, April-2019, Available at : http://www.jetir.org/papers/JETIR1904P24.pdf