Challenge Labels

The non-cortical regions of interest protocol (Table 1) follows NeuroMorphometric efforts:

The cortical parcellation protocol (Table 2) follows the brainCOLOR efforts:

Description of the Data Set

Overview

The data set is a collection of neuroanatomically labeled MRI scans. Each consists of two volumetric data sets: a 3D block of MRI data and a 3D block of labels. The labels are numbers that correspond voxel-by-voxel with the MRI data and indicate the neuroanatomical structure present at that voxel. The label numbers are an index to a list of neuroanatomical names and RGB colors.

The data sets are provided in NIFTI format (see http://nifti.nimh.nih.gov/). Specifically, the 3D volumes are provided as .nii files and the neuroanatomical structure list is an .xml file. For each scan these three files are provided as a single compressed tar file (.tgz).

The original MRI scans were obtained from the Open Access Series of Imaging Studies (OASIS) project web site (http://www.oasis-brains.org/). The specific scans used were from a sub-set of the “Crosssectional MRI Data in Young, Middle Aged, Nondemented and Demented Older Adults” called “reliability data”. Here, 20 normal control volunteer subjects were each scanned twice. They were all right handed and include 8 males and 12 females. Their ages ranged from 19 to 34 with an average of 23.4 years old.

Analysis of each Scan

Preprocessing: The processing for each individual scan begins with automated bias field inhomogeneity correction. Next, each scan is positionally normalized. Three anatomical landmarks are located: the anterior commissure (AC), the posterior commissure (PC), and a mid-sagittal point. The scan is reoriented and resliced so that anatomical labeling can be done in coronal planes that follow the AC-PC axis. Brain extent landmarks are then chosen and scans are cropped to make efficient use of computer RAM and screen area.

Outlining: The label for each voxel is determined by locating the boundaries of specific neuroanatomical regions of interest and then labeling those regions. Outlines are created and edited by highly trained Neuroanatomical Technicians (NeuroTechs) using Neuromorphometrics’ software, “NVM”. This tool provides histograms, isointensity contours, the ability to manual draw and erase borders, and many other features. NVM is designed and optimized to let the user efficiently delineate and label anatomy in 3 dimensions. The exact specification of each region of interest is defined in Neuromorpmometrics’ Segmentation Protocols version 1.1, dated 16-Jan-07. To facilitate the accurate outlining of regions, the MRI data is zoomed in to twice its normal size. Outlines are made on each slice of the MRI volume and each closed outline is assigned a neuroanatomical label corresponding to that region of interest. Any questions arising from problematic neuroanatomy are resolved by a Consulting Neuroanatomist.

Postprocessing: MRI scans are saved in the NIFTI format at the original scan size. The 3D blocks of labels are created by assigning a label number to each voxel inside an outline. Because outlining is performed on MRI scans at twice the original size, the label data is shrunken before saving in the NIFTI format.

Quality Control: The main tools for quality control are the eyes of the NeuroTechs using NVM, but additional automated and visual checks are performed throughout the analysis. Each scan is inspected by a different NeuroTech or Consulting Anatomist. Labeled regions in 3D are filled in colors for each specific neuroanatomical region so that mistakes “jump out” during visual inspection. A red-green color scheme is used to check for proper left-right labeling. Labeled regions are visualized in 3D where they can be rotated and zoomed to check for errors. Volumes of regions are calculated and compared with normal value ranges. Overlap scores are calculated between individual scans and an atlas. The outlines for any volumes or overlap scores that appear to be inappropriately large or small are checked for labeling errors and the outlines are adjusted as necessary. Since each subject was scanned twice, the volumes and overlap scores are also evaluated.

Table 1. Non-Cortical Regions of Interest

Sub-Cortical Label
3rd Ventricle
4th Ventricle
5th Ventricle
Right Accumbens Area
Left Accumbens Area
Right Amygdala
Left Amygdala
Brain Stem
Right Caudate
Left Caudate
Right Cerebellum Exterior
Left Cerebellum Exterior
Right Cerebellum White Matter
Left Cerebellum White Matter
Right Cerebral Exterior
Left Cerebral Exterior
Right Cerebral White Matter
Left Cerebral White Matter
CSF
Right Hippocampus
Left Hippocampus
Right Inf Lat Vent
Left Inf Lat Vent
Right Lateral Ventricle
Left Lateral Ventricle
Right Lesion
Left Lesion
Right Pallidum
Left Pallidum
Right Putamen
Left Putamen
Right Thalamus Proper
Left Thalamus Proper
Right Ventral DC
Left Ventral DC
Optic Chiasm
Cerebellar Vermal Lobules I-V
Cerebellar Vermal Lobules VI-VII
Cerebellar Vermal Lobules VIII-X

Table 2. Cortical regions of interest and their abbreviations.

Abbreviation Cortical Label
ACgG anterior cingulate gyrus
AIns* anterior insula
AnG angular gyrus
AOrG anterior orbital gyrus
Calc* calcarine cortex
CgG cingulate gyrus
CO central operculum
Cun cuneus
Ent entorhinal area
FO frontal operculum
FRP frontal pole
FuG fusiform gyrus
GRe gyrus rectus
IOG inferior occipital gyrus
ITG inferior temporal gyrus
LiG lingual gyrus
LOrG lateral orbital gyrus
MCgG* middle cingulate gyrus
MFC* medial frontal cortex
MFG middle frontal gyrus
MOG* middle occipital gyrus
MOrG medial orbital gyrus
MPoG postcentral gyrus, medial segment
MPrG precentral gyrus, medial segment
MSFG superior frontal gyrus, medial segment
MTG middle temporal gyrus
OCP occipital pole
OFuG* occipital fusiform gyrus
OpIFG opercular part of the inferior frontal gyrus
OrIFG orbital part of the inferior frontal gyrus
PCgG posterior cingulate gyrus
PCu precuneus
PHG parahippocampal gyrus
PIns* posterior insula
PO parietal operculum
PoG postcentral gyrus
POrG posterior orbital gyrus
PP* planum polare
PrG precentral gyrus

'* Region label and abbreviation not included in the NeuroNames database

Looking for something?

Use the form below to search the wiki:

 

Still not finding what you're looking for? Drop a comment on a post or contact us so we can take care of it!