Applications of Deep Learning to Neurodevelopment in Pediatric Imaging: Achievements and Challenges

Deep learning has achieved remarkable progress, particularly in neuroimaging analysis. applications have also been extended from adult to pediatric medical images, and thus, this paper aims present a systematic review of recent research. We first introduce the commonly used deep methods architectures neuroimaging, such as convolutional neural networks, auto-encoders, generative adversarial networks. A non-exhaustive list publicly available datasets repositories are included, followed by categorical works MRI-based studies past five years. These categorized into recognizing neurodevelopmental disorders, identifying brain tissue structures, estimating age/maturity, predicting neurodevelopment outcomes, optimizing MRI imaging Finally, we discuss achievements challenges on these neuroimaging.