Computer-aided Diagnosis systems (CAD) have attracted the attention of researchers around the world due to the great success achieved in helping healthcare specialists in the interpretation, analysis, identification and treatment of numerous existing diseases. These systems can be computational processes, models or tools which they assist the specialists through the use of database techniques, image processing, computer vision and artificial intelligence applied to different data modalities such as images, signals, videos and textual documents. In the literature, some CAD systems can achieve accuracy results similar or superior to specialists in the treatment of some diseases such as breast cancer, diabetic retinopathy and melanoma. This project aims to propose new deep learning approaches applied to different medical imaging modalities (e.g., X-Ray, MRI and Computed Tomography) and some diseases existing in the literature (e.g., breast cancer, lung cancer and glaucoma).