Computational protocol: Evaluation of a deep learning approach for the segmentation of brain tissues and white matter hyperintensities of presumed vascular origin in MRI

Similar protocols

Protocol publication

[…] The method was evaluated with images from three different data sets. For all three data sets, MR images were acquired with a Philips Achieva 3T scanner using the same acquisition protocol: a 3D T1-weighted image (TR: 7.9 ms, TE: 4.5 ms), a T1-weighted inversion recovery (IR) image (TR: 4416 ms, TE: 15 ms, TI: 400 ms), and a T2-weighted fluid attenuated inversion recovery (FLAIR) image (TR: 11,000 ms, TE: 125 ms, TI: 2800 ms) (). The 3D T1-weighted image and the T1-weighted IR image were registered to the T2-weighted FLAIR image with elastix (). After registration, all images had a voxel size of 0.96 × 0.96 × 3.0 mm3. The images were corrected for MR field bias and brain masks were generated with SPM12 (). […]

Pipeline specifications

Software tools elastix, SPM
Application Magnetic resonance imaging
Diseases Brain Diseases, Leukoencephalopathies