In recent years, deep convolutional neural networks (CNNs) have shownrecord-shattering performance in a variety of computer vision problems, such asvisual object recognition, detection and segmentation. These methods have alsobeen utilized in medical image analysis domain for lesion segmentation,anatomical segmentation and classification. We present an extensive literaturereview of CNN techniques applied in brain magnetic resonance imaging (MRI)analysis, focusing on the architectures, pre-processing, data-preparation andpost-processing strategies available in these works. The aim of this study isthree-fold. Our primary goal is to report how different CNN architectures haveevolved, now entailing state-of-the-art methods by extensive discussion of thearchitectures and examining the pros and cons of the models when evaluatingtheir performance using public datasets. Second, this paper is intended to be adetailed reference of the research activity in deep CNN for brain MRI analysis.Finally, our goal is to present a perspective on the future of CNNs, which webelieve will be among the growing approaches in brain image analysis insubsequent years.
展开▼
