CEREBRUM-7T: Fast and Fully-volumetric Brain Segmentation of 7 Tesla MR Volumes

Abstract

<jats:title>A<jats:sc>bstract</jats:sc></jats:title><jats:p>Ultra high field MRI enables sub-millimetre resolution imaging of human brain, allowing to disentangle complex functional circuits across different cortical depths. The capability of using these innovative scanners at 7 Tesla (7T) poses new challenges, as for example those related to the current lack of standardised acquisition protocols, and of automatised pipelines for image analysis. Segmentation, meant as the partition of MR brain images in multiple anatomical classes, is an essential step in many functional and structural neuroimaging studies. In this work, we design and test <jats:monospace>CEREBRUM-7T</jats:monospace>, an optimised end-to-end CNN architecture, that allows to segment a whole 7T T1<jats:sub>w</jats:sub> MRI brain volume at once, without the need of partitioning it into 2D or 3D tiles. Despite deep learning (DL) methods are recently starting to emerge in 3T literature, to the best of our knowledge, <jats:monospace>CEREBRUM-7T</jats:monospace> is the first example of DL architecture directly applied on 7T data with the purpose of segmentation. Training is performed in a weakly supervised fashion, since it exploits a ground-truth (GT) not exempt from errors. The generated model is able to produce accurate multi-structure segmentation masks on six different classes, in only few seconds. In the experimental part, a combination of objective numerical evaluations, and subjective analysis carried out by experienced neuroimaging users, confirms that the proposed solution outperforms the GT it was trained on in segmentation accuracy, and it is suitable for many neuroimaging studies. Furthermore, to allow replicability and encourage extensions, we release the code, 7T data (145 volumes), and other materials, including the GT and the survey.</jats:p>

Prospective Motion Correction and Automatic Segmentation of Penetrating Arteries in Phase Contrast MRI at 7 T

“…Although CNN-based tissue segmentation of 7 T healthy brain MRI images has been reported before [30,31], our study is the first to evaluate the performance of such an approach for WM segmentation based on T2w images. CEREBRUM-7T employed 3 layers of 3D convolutional blocks and achieved DSC of 0.9 and 0.86 in WM and BG [30], respectively, which are superior to the DSC results in Table 1. However, T 1 weighted images which have higher tissue contrast were employed.…”