Applications of deep learning techniques on biomedical images

Deep learning is an approach that includes machine learning algorithms that enable learning and classification of features for unknown data by creating models with many layered artificial neural networks on previously known data. This approach is the result of increased applicability of artificial intelligence technology and work has begun to increase. Using deep learning technologies, especially image recognition and classification studies have also accelerated. The most basic deep learning algorithms are multi-layer artificial neural networks and convolutional neural networks. These algorithms are used extensively, especially in the field of image processing. In this study, it is aimed to determine the diseased structure of biomedical images by using Convolutional Neural Networks (CNNs) method and to classify medical images by using deep learning terminology. In this study, some data from several medical data sets were trained and then the desired image (diseased area, tissue or cell) was detected by means of various deep learning network models of the data which were separated for test purposes. At the end of this study, it was seen that deep learning was successful in classifying biomedical images.