DeepNeuro: an open-source deep learning toolbox for neuroimaging
This work addresses the problem of practical implementation of deep learning in neuroimaging for clinicians and researchers, though it is incremental as it builds on existing methods by providing a specialized toolbox.
The authors tackled the challenge of translating neural networks into clinical neuroimaging practice by introducing DeepNeuro, an open-source deep learning toolbox that facilitates the design, training, and deployment of algorithms with minimal friction, including features for preprocessing, postprocessing, and packaging into shareable Docker containers and command-line interfaces.
Translating neural networks from theory to clinical practice has unique challenges, specifically in the field of neuroimaging. In this paper, we present DeepNeuro, a deep learning framework that is best-suited to putting deep learning algorithms for neuroimaging in practical usage with a minimum of friction. We show how this framework can be used to both design and train neural network architectures, as well as modify state-of-the-art architectures in a flexible and intuitive way. We display the pre- and postprocessing functions common in the medical imaging community that DeepNeuro offers to ensure consistent performance of networks across variable users, institutions, and scanners. And we show how pipelines created in DeepNeuro can be concisely packaged into shareable Docker containers and command-line interfaces using DeepNeuro's pipeline resources.