Published Paper Details:

Melanoma and other skin malignancies are among the most serious medical problems of the twenty-first century due to their difficult and subjective human interpretation and extremely expensive and complex diagnosis. When it comes to lethal illnesses like melanoma, early detection is crucial for assessing the likelihood of recovery. We think the use of automated approaches will aid in early diagnosis, particularly when a batch of photos has a variety of diagnoses. Therefore, in contrast to traditional medical personnel-based detection, we describe in this report a fully automated approach for identifying dermatological diseases using images of lesions. Our model is developed in three stages, which include data gathering and augmentation, model construction, and prediction. Convolutional Neural Networks and Support Vector Machine are two AI algorithms that we combined with image processing technologies to create a better structure and achieve an accuracy of 90%.