Skin Cancer Diagnosis and Detection Using Deep Learning

If early diagnosis and early detection of skin cancer is achieved, many patients can survive. The traditional method for the public has always suffered from problems such as imprecision and biased results. This study therefore presents a method that can be introduced in a comprehensive workflow to build a healthcare system with artificial intelligence. More exactly an intelligent system that can detect and diagnose skin cancer. While most work that use deep learning techniques focuses either on detection or diagnosis skin cancer, we have use these techniques to develope a model that does both tasks at once. Transfer learning was applied on different models of object detection and diagnosis using a dataset from Kaggle with TensorFlow API. The dataset images are a total set of 3297 dermatoscopic images. To train our model, 2637 images were used. And to test it, 660 images were used. VGG16 was used as an object recognition backbone network (for diagnosis) and Yolo as the object detection framework. The evaluation accuracy of the model was more than 83 %, which is promising.