Skin cancer detection using Machine Learning
Setlem Bhargavi, Vommi Sowmya, S Syama, S Lekshmi
Abstract
This paper addresses one of the most significant medical problems because of its challenging and subjective human interpretation. Early diagnosis is crucial for evaluating the likelihood of treatment, particularly in situations of fatal illnesses like melanoma, and it is quite expensive to diagnose. Early diagnosis will benefit from the use of automated methods, especially when a group of images contains a variety of diagnoses. As a result, this paper presents a fully automated method for diagnosing dermatological conditions from lesion data. This automated process replaces the traditional method of detection and is divided into three parts that are data collection, preprocessing, and output prediction using a support vector machine preprocessing involves removing hair and shade of the data to get better prediction.