mri_convert: orientation issue - Freesurfer

8 Sep 2011


      Hello,
I am using mri_convert to convert dicom image into analyze format using:
mri_convert -i PATH_TO_DICOM_HERE --out_type analyze -o analyze.img
The problem is:
The hdr.hist.orient in the output file is set to 0, but the pixels in
the analyze image are  saggital.
Does anybody have any suggestions as to what could be wrong?
Regards
Mohana Ramaratnam
DIAGNOSTICS:
mri_info on the dicom reports:
Orientation   : PIL
Primary Slice Direction: sagittal
and that on the analyze image reports:
Orientation   : LAS
Primary Slice Direction: axial
OUTPUT of mri_convert:
INFO: Scanning for Series Number 4
INFO: found 176 files in series
INFO: loading series header info.
INFO: sorting.
RunNo = 3
INFO: (256 256 176), nframes = 1, ismosaic=0
PE Dir ROW ROW
AutoAlign matrix detected
AutoAlign Matrix ---------------------
 0.998   0.038  -0.051  -2.542;
-0.051   0.954  -0.294  -15.875;
 0.038   0.296   0.954  -2.910;
 0.000   0.000   0.000   1.000;
FileName                FILE_NAME_HERE
Identification
        NumarisVer        syngo MR B17
        ScannerModel      TrioTim
        PatientName       65433
Date and time
        StudyDate         20110708
        StudyTime         110252.125000
        SeriesTime        111446.078000
        AcqTime           110635.592500
Acquisition parameters
        PulseSeq          tfl3d1_ns
        Protocol          t1_mpr_1mm_p2_pos50
        PhEncDir          ROW
        EchoNo            1
        FlipAngle         8
        EchoTime          3.16
        InversionTime     1000
        RepetitionTime    2400
        PhEncFOV          256
        ReadoutFOV        256
Image information
        RunNo             3
        SeriesNo          4
        ImageNo           1
        NImageRows        256
        NImageCols        256
        NFrames           1
        SliceArraylSize   1
        IsMosaic          0
        ImgPos            104.6950 154.3827 112.3121
        VolRes              1.0000   1.0000   1.0000
        VolDim            256      256      176
        Vc                 -0.0493  -0.9976   0.0480
        Vr                 -0.0403  -0.0460  -0.9981
        Vs                 -0.9980   0.0511   0.0379
        VolCenter           5.4129  25.2910  -5.9690
        TransferSyntaxUID 1.2.840.10008.1.2.1
INFO: no Siemens slice order reversal detected (good!).
TR=2400.00, TE=3.16, TI=1000.00, flip angle=8.00
i_ras = (-0.0492738, -0.997631, 0.0480037)
j_ras = (-0.0402601, -0.0460391, -0.998128)
k_ras = (-0.997974, 0.0511142, 0.0378956)
writing to 110708_TC33875.img...
Analyze Output Matrix
-0.049  -0.040  -0.998   105.782;
-0.998  -0.046   0.051   155.375;
 0.048  -0.998   0.038   113.224;
 0.000   0.000   0.000   1.000;
--------------------
No such orientation specified in Analyze7.5. Set orient to 0
Direction cosines for FILE_NAME_HERE
are:
  x_r =  -0.0493, y_r =  -0.0403, z_r =  -0.9980, c_r =     5.4129
  x_a =  -0.9976, y_a =  -0.0460, z_a =   0.0511, c_a =    25.2910
  x_s =   0.0480, y_s =  -0.9981, z_s =   0.0379, c_s =    -5.9690
INFO: set hdr.hist.orient to 'transverse unflipped'