Introduction: Recently, deep learning based methods have become an important approach
to the accurate analysis of medical images.
Materials and Methods: This paper provides a comprehensive survey of the most important deep
learning based methods that have been developed for medical image processing. A number of important
contributions made in the last five years are summarized and surveyed.
Results: Specifically, deep learning-based algorithms developed for image segmentation, image
classification, registration, object detection and other important problems are reviewed.
Conclusion: In addition, an overview of challenges that currently exist in the field and potential
directions for future research is provided at the end of the survey.