File Requirements Specification

To get best results from MIAC Brain Volumes, it is recommended that a sagittal MP-RAGE / or IR-SPGR or similar is produced.

ADNI provides suggestions for appropriate protocols for different scanners and software of General Electric (GE) Healthcare, Philips Medical Systems and Siemens Medical Solutions and for different magnetic strength (1.5 Tesla and 3 Tesla).

Briefly, image packages should be T1-weighted, sagittal, non-contrast enhanced, high resolution 3D MR series. High resolution means that the edge of the voxels are (1.2 mm slice thickness, Range 1-1.5 mm). Folding artifacts or moving artifacts should be avoided. Full head coverage is an absolute requirement. The applied field strength may be 1.5 or 3 Tesla as available. Applicable sequences are SPGR for GE scanners and MPRAGE for Siemens scanners or similar sequences for other scanners.

Parameters that are common to different scanning protocols are listed in following table:

Attribute Essential Parameters (Range)
Slice Orientation Sagittal
Mode 3D
slice thickness [mm] 1.2 (range 1.0 - 1.5)
Number of Slices 120-180
Field of view (frequency x phase) [mm] 240 x 240 (max 256)
Image matrix (frequency x phase) 192 x 192
Resulting In plane resolution [mm] 1.25 x 1.25
Slice thickness 1.2 mm (1 - 1.5 mm)
Slice gap 0 mm

Expected File format

MIAC Brain Volumes expects a single file in zip format containing a series of DICOM files. Files, which are not DICOM or not MRI images are ignored. The archive may not contain image files from multiple series and may not contain images of variegated size.

Encoding: the input scan should be encoded using one of the following DICOM transfer syntaxes:

  • Implicit VR Little Endian Default Transfer Syntax
  • Explicit VR Little Endian
  • Deflated Explicit VR Little Endian
  • Explicit VR Big Endian
  • JPEG Lossless, Non-Hierarchical (Process 14)
  • JPEG Lossless, Non-Hierarchical, First-Order Prediction (Process 14 [Selection Value 1])
  • JPEG 2000 Image Compression (Lossless Only)
  • JPEG-LS Lossless Image Compression