Automatic segmentation of MR brain images with a convolutional neural network.Accurate automatic brain image segmentation in magnetic resonance (MR) Especially the field of neonatal brain image segmentation has developed rapidly in the last ten years.

A popular approach for brain image segmentation is the use of (population specific) atlases,

To obtain anatomically correct segmentations, these methods need explicit spatial and intensity information.For atlas-based methods spatial information is provided in the form of an atlas which is deformed to match the image at hand.In recent years, CNNs also gained popularity in the field of medical image analysis. In contrast to classical machine learning methods, CNNs do not require a set of hand-crafted features for the classification,

learn sets of convolution kernels that are specifically trained for the classification problem at hand. While classical machine learning methods applied to image segmentation would use e.g.Gaussian or Haar-like kernels to acquire appearance information, CNNs optimise sets of kernels based on the provided training data.In this way, the system can automatically extract information that is relevant for the task. CNNs have also been used for brain image segmentation.

Intensity information is, in pattern recognition methods, included as a set of features based on (local) intensity,

atlas-based methods are typically performed by matching intensity information between the atlas and target images.

For methods based on pattern recognition spatial information is included in the feature set as distances within an atlas space, distances to a brain mask, probabilistic results of a previous segmentation step, or by imposing anatomical constraints.

To obtain anatomically correct segmentations, these methods need explicit spatial and intensity information.