Increasing the quality of melanoma dermatoscopic images using the structure of deep learning

In the image-based diagnosis of diseases, good-quality of images enhances the accuracy of diagnosis. Skin cancers are one of the most common cancers in the human body and Melanoma is one of the most dangerous types of it. diagnosis in the early stages reduces the mortality caused by this fatal cancer. Therefore, the diagnosis of benign or malignant and the tumor location is very important in medicine. Dermatoscopes are non-invasive means of photographing skin damage. Today, the volume of data available in images allows us to create useful tools for extracting information from images and video sequences so that we can categorize or analyze content without human intervention. Artificial intelligence technology has grown dramatically in recent years and the tools used in this technology have the potential to improve the quality of low-resolution images. By improving the quality and improving the visualization of small tissues in dermatoscopic images, the final diagnosis of benign or malignant mass will be more accurate. Nowadays, deep learning techniques for medical image analysis have provided the possibility for the development of medical imaging-based intelligent diagnostic systems that can assist the expert in making better decisions about patients' health.In this article deep learning has been used to enhance the quality of dermatoscopic images and will also improve the quality of the final image that we measure this quality with the PSNR criterion. PSNR output is the recommended method for several standard ۴۶.۰۷ images, while the best output of previous methods on these images is ۴۴.۰۲.

کلیدواژه ها:

Skin Cancer ، Melanoma ، Deep Learning ، Artificial Intelligence ، Dermatoscopic Images