Deep learning is a subcategory of machine learning and artificial intelligence that autonomously learns from data to determine the rules for a desired task. This thesis explores how deep learning can be applied in image analysis and medical diagnosis, outperforming standard algorithmic methods and simpler machine-learning methods. Examples of applications include particle tracking and characterization in 2D and 3D, segmentation, characterization and counting of biological cells, image transformation, and the diagnosis of a genetic disease and prediction of short- and long-term morbidity in patients with congenital-heart-disease.
