Deep Learning in Medical Imaging: General Overview

摘要: The artificial neural network (ANN)-a machine learning technique inspired by the human neuronal synapse system-was introduced in 1950s. However, ANN was previously limited its ability to solve actual problems, due vanishing gradient and overfitting problems with training of deep architecture, lack computing power, primarily absence sufficient data train computer system. Interest this concept has lately resurfaced, availability big data, enhanced power current graphics processing units, novel algorithms network. Recent studies on technology suggest potentially perform better than humans some visual auditory recognition tasks, which may portend applications medicine healthcare, especially medical imaging, foreseeable future. This review article offers perspectives history, development, technology, particularly regarding imaging.