English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Neuroanatomical segmentation in magnetic resonance imaging (MRI) of the brainis a prerequisite for volume, thickness and shape measurements. This workintroduces a new highly accurate and versatile method based on 3D convolutionalneural networks for the automatic segmentation of neuroanatomy in T1-weightedMRI. In combination with a deep 3D fully convolutional architecture, efficientlinear registration-derived spatial priors are used to incorporate additionalspatial context into the network. An aggressive data augmentation scheme usingrandom elastic deformations is also used to regularize the networks, allowingfor excellent performance even in cases where only limited labelled trainingdata are available. Applied to hippocampus segmentation in an elderlypopulation (mean Dice coefficient = 92.1%) and sub-cortical segmentation in ahealthy adult population (mean Dice coefficient = 89.5%), we demonstrate newstate-of-the-art accuracies and a high robustness to outliers with the samearchitecture. Further validation on a multi-structure segmentation task in ascan-rescan dataset demonstrates accuracy (mean Dice coefficient = 86.6%)similar to the scan-rescan reliability of expert manual segmentations (meanDice coefficient = 86.9%), and improved reliability compared to both expertmanual segmentations and automated segmentations using FIRST. Furthermore, ourmethod maintains a highly competitive runtime performance (e.g. requiring only10 seconds for left/right hippocampal segmentation in 1x1x1 MNI stereotaxicspace), orders of magnitude faster than conventional multi-atlas segmentationmethods.

Accurate and robust segmentation of neuroanatomy in T1-weighted MRI by combining spatial priors with deep convolutional neural networks