Computational protocol: Multiclass Classification for the Differential Diagnosis on the ADHD Subtypes Using Recursive Feature Elimination and Hierarchical Extreme Learning Machine: Structural MRI Study

Similar protocols

Protocol publication

[…] The raw data on their own provide very little information; therefore, preprocessing is required to extract the patterns of useful knowledge []. All of the structural MRI images were preprocessed using the fully automated pipeline of FreeSurfer 5.3.0 [–] for volumetric segmentation and cortical reconstruction. The software proceeds with motion correction, T1-weighted image averaging, and registration to the Talairach space, followed by skull stripping with a deformable template model. The white and pial surface was generated for each hemisphere. A cortical surface-based atlas (DKT atlas) was mapped to a sphere aligning the cortical folding patterns, which provided accurate matching of the morphologically homologous cortical locations across subjects. For each of the DKT31 protocol-based segments, Freesurfer calculated nine different measures, including, surface vertices, surface area, gray matter volume, average cortical thickness, cortical thickness standard deviation, cortical mean curvature, cortical Gaussian curvature, cortical folding index, and cortical curvature indices []. We used five of the above-mentioned measures in this study. The surface area was calculated by computing the area of every triangle in a standardized spherical surface tessellation. The average shortest distance between white and pial surfaces denotes the cortical thickness at each vertex of the cortex. The local curvature was computed using the registration surface based on the folding patterns. The folding index over the whole cortical surface was measured for each subject [].After preprocessing, the subcortical regions were masked to separate the significant cortical data by using AFNI. Finally, the cortical thickness data were converted into surface maps using the AFNI program MapIcosahedron. These maps were then subjected to heat-kernel-based smoothing using the AFNI program SurfSmooth with a 30-mm kernel. […]

Pipeline specifications

Software tools FreeSurfer, AFNI
Applications Magnetic resonance imaging, Functional magnetic resonance imaging
Organisms Homo sapiens
Diseases Brain Diseases
Chemicals Hydrogen