MRI Scanners Sequences Tested

We need to ensure that we are extracting the correct header information of the DICOM files. This page also lists different data acquisition system that produced data for testing the pipeline.

Nibabel so far only supports a limited number of DICOM formats to extract header information adequately. Important information are: bValue(s), number of b0, number of directions, gradient list

Lausanne

...

Boston

• MGH SIEMENS TrioTim AX DIFFUSION

Details:

nibabel produced a gradient table for this dataset that did not match expectation. Running 'mri_info' from freesurfer on this dataset yielded the following info below. This data was processed using the gradient table referenced in 'GradFile' below:

Getting Series No
Scanning Directory 
INFO: loading series header info.
INFO: sorting.
INFO: (112 112  79), nframes = 35, ismosaic=1
Could not parse NUMARIS version string syngo MR B15
found in dicom tag 18,1020 (len = 3 != 6)
Repetition Time = 10261.7, TR = 10261.7 ms
PE Dir COL COL
AutoAlign matrix detected 
AutoAlign Matrix --------------------- 
 1.000   0.000   0.000   0.000;
 0.000   1.000   0.000   0.000;
 0.000   0.000   1.000   0.000;
 0.000   0.000   0.000   1.000;

This looks like an MGH DTI volume
MGH DTI SeqPack Info
0 ep_b0#1  0.000000 1
1 ep_b0#2  0.000000 2
2 ep_b0#3  0.000000 3
3 ep_b0#4  0.000000 4
4 ep_b0#5  0.000000 5
5 ep_b1000#1 1000.000000 1
6 ep_b1000#2 1000.000000 2
7 ep_b1000#3 1000.000000 3
8 ep_b1000#4 1000.000000 4
9 ep_b1000#5 1000.000000 5
10 ep_b1000#6 1000.000000 6
11 ep_b1000#7 1000.000000 7
12 ep_b1000#8 1000.000000 8
13 ep_b1000#9 1000.000000 9
14 ep_b1000#10  1000.000000 10
15 ep_b1000#11  1000.000000 11
16 ep_b1000#12  1000.000000 12
17 ep_b1000#13  1000.000000 13
18 ep_b1000#14  1000.000000 14
19 ep_b1000#15  1000.000000 15
20 ep_b1000#16  1000.000000 16
21 ep_b1000#17  1000.000000 17
22 ep_b1000#18  1000.000000 18
23 ep_b1000#19  1000.000000 19
24 ep_b1000#20  1000.000000 20
25 ep_b1000#21  1000.000000 21
26 ep_b1000#22  1000.000000 22
27 ep_b1000#23  1000.000000 23
28 ep_b1000#24  1000.000000 24
29 ep_b1000#25  1000.000000 25
30 ep_b1000#26  1000.000000 26
31 ep_b1000#27  1000.000000 27
32 ep_b1000#28  1000.000000 28
33 ep_b1000#29  1000.000000 29
34 ep_b1000#30  1000.000000 30
bValue = 1000
nB0 = 5
nDir = 30
GradFile /chb/freesurfer/x86_64-Darwin/packages/freesurfer/dev/diffusion/mgh-dti-seqpack/gradient_mgh_dti30.gdt
Volume information for 218000-000011-000001.dcm
          type: siemens_dicom
    dimensions: 112 x 112 x 79
   voxel sizes: 1.9643, 1.9643, 2.0000
          type: SHORT (4)
           fov: 112.000
           dof: 0
        xstart: -56.0, xend: 56.0
        ystart: -56.0, yend: 56.0
        zstart: -39.5, zend: 39.0
            TR: 10261.70 msec, TE: 91.00 msec, TI: -1.00 msec, flip angle: 90.00 degrees
       nframes: 1
       PhEncDir: COL
ras xform present
    xform info: x_r =  -1.0000, y_r =  -0.0000, z_r =  -0.0006, c_r =     1.6127
              : x_a =  -0.0000, y_a =  -0.9990, z_a =   0.0448, c_a =    11.4873
              : x_s =  -0.0006, y_s =   0.0448, z_s =   0.9990, c_s =    30.1320
Orientation   : LPS
Primary Slice Direction: axial

voxel to ras transform:
               -1.9643  -0.0000  -0.0012   111.6590
               -0.0001  -1.9623   0.0896   117.8423
               -0.0011   0.0880   1.9980   -53.6501
                0.0000   0.0000   0.0000     1.0000

voxel-to-ras determinant 7.71684

ras to voxel transform:
               -0.5091  -0.0000  -0.0003    56.8301
               -0.0000  -0.5086   0.0228    61.1553
               -0.0003   0.0224   0.4995    24.1924
                0.0000   0.0000   0.0000     1.0000