Siemens phase encoding direction mismatch between dicom header and pulse sequence from scanner

[estephanmoana]

Hello, this is probably not an issue with dcm2niix but it is arising from dicom to nifti conversion and I am stuck. I used heudiconv_v0.11.3 singularity container to convert legacy data from a Siemens 3T Trio, and I noticed that the phase encoding direction reported in the json files do not match what is described in the pulse sequence printout from the scanner. In fact, the phase encoding direction seems to vary from one BOLD run to another, when they should have the same imaging parameters. Checking the dicom headers, it seems that they indeed have different phase encoding directions reported.

I am hoping that you could point me in the right direction. Should I disregard the DICOM header info and trust the pulse sequence printout from the scanner?

Below I pasted some of the information about the phase encoding direction in example BOLD runs. Screenshots of the pulse sequence are in the end. Thank you.

frun_stimBOLD_auditory_01, DICOM header

cn2107:~ moana004$ cd '/home/moanae/moana004/project_FM_Gracely/data_rawmri_dicom/HC0003_20130110/func/frun_stimBOLD_auditory_01' cn2107:frun_stimBOLD_auditory_01 moana004$ cn2107:frun_stimBOLD_auditory_01 moana004$ dicom_firstfile=ls | head -1; dicom_hdr ${dicom_firstfile} DICOM File: 1.3.12.2.1107.5.2.32.35412.2013011012523710876311403.dcm Object type: ELEMENT LIST Object size: 508856 Group: 0002, Length: 166 0002 0000 4 [140 ] // META Group Length// 166 0002 0001 2 [156 ] //META File Meta Information Version//hex OB: (len 2)00 01 0002 0002 26 [166 ] //META Media Stored SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0002 0003 52 [200 ] //META Media Stored SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012523710876311403 0002 0010 20 [260 ] // META Transfer Syntax UID//1.2.840.10008.1.2.1 0002 0012 26 [288 ] // META Implementation Class UID//1.2.826.0.1.3680043.2.737 Group: 0008, Length: 916 0008 0005 10 [322 ] // ID Specific Character Set//ISO_IR 100 0008 0008 28 [340 ] // ID Image Type//ORIGINAL\PRIMARY\M\ND\MOSAIC 0008 0012 8 [376 ] // ID Instance Creation Date//20130110 0008 0013 14 [392 ] // ID Instance Creation Time//125351.390000 0008 0016 26 [414 ] // ID SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 0018 52 [448 ] // ID SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012523710876311403 0008 0020 8 [508 ] // ID Study Date//20130110 0008 0021 8 [524 ] // ID Series Date//20130110 0008 0022 8 [540 ] // ID Acquisition Date//20130110 0008 0023 8 [556 ] // ID Image Date//20130110 0008 0030 14 [572 ] // ID Study Time//115036.625000 0008 0031 14 [594 ] // ID Series Time//125351.390000 0008 0032 14 [616 ] // ID Acquisition Time//125232.330000 0008 0033 14 [638 ] // ID Image Time//125351.390000 0008 0050 0 [660 ] // ID Accession Number// 0008 0060 2 [668 ] // ID Modality//MR 0008 0070 8 [678 ] // ID Manufacturer//SIEMENS 0008 0080 28 [694 ] // ID Institution Name//University of North Carolina 0008 0081 48 [730 ] // ID Institution Address//Mason Farm Rd 106,Chapel Hill,District,US,27599 0008 0090 16 [786 ] // ID Referring Physician's Name//Gracely, Richard 0008 1010 6 [810 ] // ID Station Name//MEDPC 0008 1030 10 [824 ] // ID Study Description//TRIO^HEAD 0008 103e 20 [842 ] // ID Series Description//BOLD auditory 3x3x3 0008 1040 10 [870 ] //ID Institutional Department Name//Department 0008 1050 0 [888 ] // ID Performing Physician's Name// 0008 1070 4 [896 ] // ID Operator's Name//hgm 0008 1090 8 [908 ] // ID Manufacturer Model Name//TrioTim 0008 1140 306 [0 ] // ID Referenced Image Sequence//SEQUENCE DCM Dump SEQUENCE{ Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [944 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [978 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003250863100014 DCM Dump Elements Complete Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [1046 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [1080 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003538766000018 DCM Dump Elements Complete Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [1148 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [1182 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003826830800022 DCM Dump Elements Complete DCM Dump SEQUENCE Complete} Group: 0010, Length: 120 0010 0010 6 [1242 ] // PAT Patient Name//HC0003 0010 0020 18 [1256 ] // PAT Patient ID//HC0003 10Jan2013 0010 0030 8 [1282 ] // PAT Patient Birthdate//19640211 0010 0040 2 [1298 ] // PAT Patient Sex//F 0010 1010 4 [1308 ] // PAT Patient Age//048Y 0010 1020 12 [1320 ] // PAT Patient Size//1.7526035075 0010 1030 14 [1340 ] // PAT Patient Weight//71.6676035766 Group: 0018, Length: 408 0018 0020 2 [1362 ] // ACQ Scanning Sequence//EP 0018 0021 2 [1372 ] // ACQ Sequence Variant//SK 0018 0022 2 [1382 ] // ACQ Scan Options//FS 0018 0023 2 [1392 ] // ACQ MR Acquisition Type //2D 0018 0024 12 [1402 ] // ACQ Sequence Name//*epfid2d1_64 0018 0025 2 [1422 ] // ACQ Angio Flag//N 0018 0050 2 [1432 ] // ACQ Slice Thickness//3 0018 0080 4 [1442 ] // ACQ Repetition Time//3000 0018 0081 2 [1454 ] // ACQ Echo Time//30 0018 0083 2 [1464 ] // ACQ Number of Averages//1 0018 0084 10 [1474 ] // ACQ Imaging Frequency//123.195767 0018 0085 2 [1492 ] // ACQ Imaged Nucleus//1H 0018 0086 2 [1502 ] // ACQ Echo Number//1 0018 0087 2 [1512 ] // ACQ Magnetic Field Strength//3 0018 0088 16 [1522 ] // ACQ Spacing Between Slices//2.9999999372335 0018 0089 2 [1546 ] //ACQ Number of Phase Encoding Steps//63 0018 0091 2 [1556 ] // ACQ Echo Train Length//1 0018 0093 4 [1566 ] // ACQ Percent Sampling//100 0018 0094 4 [1578 ] //ACQ Percent Phase Field of View//100 0018 0095 4 [1590 ] // ACQ Pixel Bandwidth//2298 0018 1000 6 [1602 ] // ACQ Device Serial Number//35412 0018 1020 12 [1616 ] // ACQ Software Version//syngo MR B17 0018 1030 20 [1636 ] // ACQ Protocol Name//BOLD auditory 3x3x3 0018 1251 4 [1664 ] // ACQ Transmitting Coil//Body 0018 1310 8 [1676 ] // ACQ Acquisition Matrix// 0 64 64 0 0018 1312 4 [1692 ] // ACQ Phase Encoding Direction//ROW 0018 1314 2 [1704 ] // ACQ Flip Angle//77 0018 1315 2 [1714 ] // ACQ Variable Flip Angle//N 0018 1316 16 [1724 ] // ACQ SAR//0.11310099243148 0018 1318 2 [1748 ] // ACQ DB/DT//0 0018 5100 4 [1758 ] // ACQ Patient Position//HFS Group: 0019, Length: 636 0019 0010 18 [1770 ] // //SIEMENS MR HEADER 0019 1008 12 [1796 ] // //IMAGE NUM 4 0019 1009 4 [1816 ] // //1.0 0019 100a 2 [1828 ] // // 49 0019 100b 4 [1838 ] // //17.5 0019 100f 4 [1850 ] // //Fast 0019 1011 2 [1862 ] // //No 0019 1012 12 [1872 ] // // 0 0019 1013 12 [1892 ] // // 0 0019 1014 6 [1912 ] // //0\0\0 0019 1015 24 [1926 ] // //Unimplemented 0019 1016 2 [1958 ] // //0 0019 1017 2 [1968 ] // //1 0019 1018 4 [1978 ] // //3400 0019 1028 8 [1990 ] // //Unimplemented 0019 1029 392 [2006 ] // //Unimplemented Group: 0020, Length: 426 0020 000d 56 [2406 ] // REL Study Instance UID//1.3.12.2.1107.5.2.32.35412.30000013011016503665600000001 0020 000e 58 [2470 ] // REL Series Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012514113598711199.0.0.0 0020 0010 2 [2536 ] // REL Study ID//1 0020 0011 2 [2546 ] // REL Series Number//14 0020 0012 2 [2556 ] // REL Acquisition Number//1 0020 0013 2 [2566 ] // REL Instance Number//1 0020 0032 48 [2576 ] // REL Image Position Patient//-690.8701331441\-658.92121418714\68.279913136234 0020 0037 100 [2632 ] //REL Image Orientation (Patient)//0.9992127683078\-0.030876501961\0.02490954189317\0.03489328082073\0.98276998923205\-0.1815092482999 0020 0052 52 [2740 ] // REL Frame of Reference UID//1.3.12.2.1107.5.2.32.35412.1.20130110115920187.0.0.0 0020 1040 0 [2800 ] //REL Position Reference Indicator// 0020 1041 16 [2808 ] // REL Slice Location//-39.913821776502 Group: 0028, Length: 160 0028 0002 2 [2832 ] // IMG Samples Per Pixel// 1 0028 0004 12 [2842 ] // IMG Photometric Interpretation//MONOCHROME2 0028 0010 2 [2862 ] // IMG Rows// 448 0028 0011 2 [2872 ] // IMG Columns// 448 0028 0030 4 [2882 ] // IMG Pixel Spacing//3\3 0028 0100 2 [2894 ] // IMG Bits Allocated// 16 0028 0101 2 [2904 ] // IMG Bits Stored// 12 0028 0102 2 [2914 ] // IMG High Bit// 11 0028 0103 2 [2924 ] // IMG Pixel Representation// 0 0028 0106 2 [2934 ] // IMG Smallest Image Pixel Value// 0 0028 0107 2 [2944 ] // IMG Largest Image Pixel Value// 2539 0028 1050 4 [2954 ] // IMG Window Center//762 0028 1051 4 [2966 ] // IMG Window Width//1721 0028 1055 6 [2978 ] //IMG Window Center & Width Explanation//Algo1 Group: 0029, Length: 104090 0029 0010 18 [2992 ] // //SIEMENS CSA HEADER 0029 0011 22 [3018 ] // //SIEMENS MEDCOM HEADER2 0029 1008 12 [3048 ] // //IMAGE NUM 4 0029 1009 8 [3068 ] // //20130110 0029 1010 11524 [3088 ] // //hex OB: (len 8)53 56 31 30 04 03 02 01 0029 1018 2 [14620 ] // //MR 0029 1019 8 [14630 ] // //20130110 0029 1020 92420 [14650 ] // //hex OB: (len 8)53 56 31 30 04 03 02 01 0029 1160 4 [107078 ] // //com Group: 0032, Length: 18 0032 1060 10 [107090 ] //SDY Requested Procedure Description//TRIO HEAD Group: 0040, Length: 80 0040 0244 8 [107108 ] // PRC PPS Start Date//20130110 0040 0245 14 [107124 ] // PRC PPS Start Time//115038.781000 0040 0253 16 [107146 ] // PRC PPS ID//MR20130110115038 0040 0254 10 [107170 ] // PRC PPS Description//TRIO^HEAD Group: 0051, Length: 256 0051 0010 18 [107188 ] // //SIEMENS MR HEADER 0051 1008 12 [107214 ] // //IMAGE NUM 4 0051 1009 4 [107234 ] // //1.0 0051 100a 8 [107246 ] // //TA 00.01 0051 100b 6 [107262 ] // //64*64s 0051 100c 14 [107276 ] // //FoV 1344*1344 0051 100e 24 [107298 ] // //Tra>Cor(-10.5)>Sag(1.1) 0051 100f 10 [107330 ] // //t:HEA;HEP 0051 1011 2 [107348 ] // //p2 0051 1012 4 [107358 ] // //TP 0 0051 1013 4 [107370 ] // //+LPH 0051 1015 4 [107382 ] // //D/R 0051 1016 14 [107394 ] // //p2 M/ND/MOSAIC 0051 1017 6 [107416 ] // //SL 3.0 0051 1019 6 [107430 ] // //A1/FS Group: 7fe0, Length: 401416 7fe0 0010 401408 [107448 ] // PXL Pixel Data//Data on disk DCM Dump Elements Complete Pixel array offset = 107448 (bytes) Pixel array length = 401408 (bytes) frun_stimBOLD_auditory_01, fslhd output `cn2107:~ moana004$ fslhd '/home/moanae/moana004/project_FM_Gracely/BIDS_output/sub-HC0003/func/sub-HC0003_task-auditory_run-03_bold.nii.gz' filename /home/moanae/moana004/project_FM_Gracely/BIDS_output/sub-HC0003/func/sub-HC0003_task-auditory_run-03_bold.nii.gz

sizeof_hdr 348 data_type INT16 dim0 4 dim1 64 dim2 64 dim3 49 dim4 208 dim5 1 dim6 1 dim7 1 vox_units mm time_units s datatype 4 nbyper 2 bitpix 16 pixdim0 -1.000000 pixdim1 3.000000 pixdim2 3.000000 pixdim3 3.000000 pixdim4 3.000000 pixdim5 0.000000 pixdim6 0.000000 pixdim7 0.000000 vox_offset 352 cal_max 0.000000 cal_min 0.000000 scl_slope 1.000000 scl_inter 0.000000 phase_dim 1 freq_dim 2 slice_dim 3 slice_name sequential_increasing slice_code 1 slice_start 0 slice_end 0 slice_duration 0.000000 toffset 0.000000 intent Unknown intent_code 0 intent_name intent_p1 0.000000 intent_p2 0.000000 intent_p3 0.000000 qform_name Scanner Anat qform_code 1 qto_xyz:1 -2.997638 0.104681 0.056639 88.630211 qto_xyz:2 0.092629 2.948310 -0.546707 -75.112961 qto_xyz:3 0.074739 0.544528 2.949221 -56.226753 qto_xyz:4 0.000000 0.000000 0.000000 1.000000 qform_xorient Right-to-Left qform_yorient Posterior-to-Anterior qform_zorient Inferior-to-Superior sform_name Scanner Anat sform_code 1 sto_xyz:1 -2.997638 0.104680 0.056628 88.630211 sto_xyz:2 0.092630 2.948310 -0.546707 -75.112961 sto_xyz:3 0.074729 0.544528 2.949221 -56.226753 sto_xyz:4 0.000000 0.000000 0.000000 1.000000 sform_xorient Right-to-Left sform_yorient Posterior-to-Anterior sform_zorient Inferior-to-Superior file_type NIFTI-1+ file_code 1 descrip TE=30;Time=125232.330;phase=1 aux_file
` frun_stimBOLD_mechanical_01, DICOM header

cn2107:frun_stimBOLD_auditory_01 moana004$ cd '/home/moanae/moana004/project_FM_Gracely/data_rawmri_dicom/HC0003_20130110/func/frun_stimBOLD_mechanical_01' cn2107:frun_stimBOLD_mechanical_01 moana004$ dicom_firstfile=ls | head -1; dicom_hdr ${dicom_firstfile} DICOM File: 1.3.12.2.1107.5.2.32.35412.2013011012353247793869768.dcm Object type: ELEMENT LIST Object size: 508962 Group: 0002, Length: 166 0002 0000 4 [140 ] // META Group Length// 166 0002 0001 2 [156 ] //META File Meta Information Version//hex OB: (len 2)00 01 0002 0002 26 [166 ] //META Media Stored SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0002 0003 52 [200 ] //META Media Stored SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012353247793869768 0002 0010 20 [260 ] // META Transfer Syntax UID//1.2.840.10008.1.2.1 0002 0012 26 [288 ] // META Implementation Class UID//1.2.826.0.1.3680043.2.737 Group: 0008, Length: 918 0008 0005 10 [322 ] // ID Specific Character Set//ISO_IR 100 0008 0008 28 [340 ] // ID Image Type//ORIGINAL\PRIMARY\M\ND\MOSAIC 0008 0012 8 [376 ] // ID Instance Creation Date//20130110 0008 0013 14 [392 ] // ID Instance Creation Time//123638.359000 0008 0016 26 [414 ] // ID SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 0018 52 [448 ] // ID SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012353247793869768 0008 0020 8 [508 ] // ID Study Date//20130110 0008 0021 8 [524 ] // ID Series Date//20130110 0008 0022 8 [540 ] // ID Acquisition Date//20130110 0008 0023 8 [556 ] // ID Image Date//20130110 0008 0030 14 [572 ] // ID Study Time//115036.625000 0008 0031 14 [594 ] // ID Series Time//123638.343000 0008 0032 14 [616 ] // ID Acquisition Time//123528.355000 0008 0033 14 [638 ] // ID Image Time//123638.359000 0008 0050 0 [660 ] // ID Accession Number// 0008 0060 2 [668 ] // ID Modality//MR 0008 0070 8 [678 ] // ID Manufacturer//SIEMENS 0008 0080 28 [694 ] // ID Institution Name//University of North Carolina 0008 0081 48 [730 ] // ID Institution Address//Mason Farm Rd 106,Chapel Hill,District,US,27599 0008 0090 16 [786 ] // ID Referring Physician's Name//Gracely, Richard 0008 1010 6 [810 ] // ID Station Name//MEDPC 0008 1030 10 [824 ] // ID Study Description//TRIO^HEAD 0008 103e 22 [842 ] // ID Series Description//BOLD mechanical 3x3x3 0008 1040 10 [872 ] //ID Institutional Department Name//Department 0008 1050 0 [890 ] // ID Performing Physician's Name// 0008 1070 4 [898 ] // ID Operator's Name//hgm 0008 1090 8 [910 ] // ID Manufacturer Model Name//TrioTim 0008 1140 306 [0 ] // ID Referenced Image Sequence//SEQUENCE DCM Dump SEQUENCE{ Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [946 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [980 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003250863100014 DCM Dump Elements Complete Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [1048 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [1082 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003538766000018 DCM Dump Elements Complete Object type: ELEMENT LIST Object size: 94 Group: 0008, Length: 94 0008 1150 26 [1150 ] // ID Referenced SOP Class UID//1.2.840.10008.5.1.4.1.1.4 0008 1155 52 [1184 ] // ID Referenced SOP Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012003826830800022 DCM Dump Elements Complete DCM Dump SEQUENCE Complete} Group: 0010, Length: 120 0010 0010 6 [1244 ] // PAT Patient Name//HC0003 0010 0020 18 [1258 ] // PAT Patient ID//HC0003 10Jan2013 0010 0030 8 [1284 ] // PAT Patient Birthdate//19640211 0010 0040 2 [1300 ] // PAT Patient Sex//F 0010 1010 4 [1310 ] // PAT Patient Age//048Y 0010 1020 12 [1322 ] // PAT Patient Size//1.7526035075 0010 1030 14 [1342 ] // PAT Patient Weight//71.6676035766 Group: 0018, Length: 410 0018 0020 2 [1364 ] // ACQ Scanning Sequence//EP 0018 0021 2 [1374 ] // ACQ Sequence Variant//SK 0018 0022 2 [1384 ] // ACQ Scan Options//FS 0018 0023 2 [1394 ] // ACQ MR Acquisition Type //2D 0018 0024 12 [1404 ] // ACQ Sequence Name//*epfid2d1_64 0018 0025 2 [1424 ] // ACQ Angio Flag//N 0018 0050 2 [1434 ] // ACQ Slice Thickness//3 0018 0080 4 [1444 ] // ACQ Repetition Time//3000 0018 0081 2 [1456 ] // ACQ Echo Time//30 0018 0083 2 [1466 ] // ACQ Number of Averages//1 0018 0084 10 [1476 ] // ACQ Imaging Frequency//123.195759 0018 0085 2 [1494 ] // ACQ Imaged Nucleus//1H 0018 0086 2 [1504 ] // ACQ Echo Number//1 0018 0087 2 [1514 ] // ACQ Magnetic Field Strength//3 0018 0088 16 [1524 ] // ACQ Spacing Between Slices//2.9999999372335 0018 0089 2 [1548 ] //ACQ Number of Phase Encoding Steps//63 0018 0091 2 [1558 ] // ACQ Echo Train Length//1 0018 0093 4 [1568 ] // ACQ Percent Sampling//100 0018 0094 4 [1580 ] //ACQ Percent Phase Field of View//100 0018 0095 4 [1592 ] // ACQ Pixel Bandwidth//2298 0018 1000 6 [1604 ] // ACQ Device Serial Number//35412 0018 1020 12 [1618 ] // ACQ Software Version//syngo MR B17 0018 1030 22 [1638 ] // ACQ Protocol Name//BOLD mechanical 3x3x3 0018 1251 4 [1668 ] // ACQ Transmitting Coil//Body 0018 1310 8 [1680 ] // ACQ Acquisition Matrix// 64 0 0 64 0018 1312 4 [1696 ] // ACQ Phase Encoding Direction//COL 0018 1314 2 [1708 ] // ACQ Flip Angle//77 0018 1315 2 [1718 ] // ACQ Variable Flip Angle//N 0018 1316 16 [1728 ] // ACQ SAR//0.11310099243148 0018 1318 2 [1752 ] // ACQ DB/DT//0 0018 5100 4 [1762 ] // ACQ Patient Position//HFS Group: 0019, Length: 634 0019 0010 18 [1774 ] // //SIEMENS MR HEADER 0019 1008 12 [1800 ] // //IMAGE NUM 4 0019 1009 4 [1820 ] // //1.0 0019 100a 2 [1832 ] // // 49 0019 100b 2 [1842 ] // //15 0019 100f 4 [1852 ] // //Fast 0019 1011 2 [1864 ] // //No 0019 1012 12 [1874 ] // // 0 0019 1013 12 [1894 ] // // 0 0019 1014 6 [1914 ] // //0\0\0 0019 1015 24 [1928 ] // //Unimplemented 0019 1016 2 [1960 ] // //0 0019 1017 2 [1970 ] // //1 0019 1018 4 [1980 ] // //3400 0019 1028 8 [1992 ] // //Unimplemented 0019 1029 392 [2008 ] // //Unimplemented Group: 0020, Length: 430 0020 000d 56 [2408 ] // REL Study Instance UID//1.3.12.2.1107.5.2.32.35412.30000013011016503665600000001 0020 000e 58 [2472 ] // REL Series Instance UID//1.3.12.2.1107.5.2.32.35412.2013011012345343546169566.0.0.0 0020 0010 2 [2538 ] // REL Study ID//1 0020 0011 2 [2548 ] // REL Series Number//12 0020 0012 2 [2558 ] // REL Acquisition Number//1 0020 0013 2 [2568 ] // REL Instance Number//1 0020 0032 50 [2578 ] // REL Image Position Patient//-690.87013056468\-658.92121723566\68.279911638939 0020 0037 102 [2636 ] //REL Image Orientation (Patient)//0.99921276844443\-0.0308764981072\0.02490954118898\0.03489327684568\0.98276998991476\-0.1815092453676 0020 0052 52 [2746 ] // REL Frame of Reference UID//1.3.12.2.1107.5.2.32.35412.1.20130110115920187.0.0.0 0020 1040 0 [2806 ] //REL Position Reference Indicator// 0020 1041 16 [2814 ] // REL Slice Location//-39.913821739334 Group: 0028, Length: 160 0028 0002 2 [2838 ] // IMG Samples Per Pixel// 1 0028 0004 12 [2848 ] // IMG Photometric Interpretation//MONOCHROME2 0028 0010 2 [2868 ] // IMG Rows// 448 0028 0011 2 [2878 ] // IMG Columns// 448 0028 0030 4 [2888 ] // IMG Pixel Spacing//3\3 0028 0100 2 [2900 ] // IMG Bits Allocated// 16 0028 0101 2 [2910 ] // IMG Bits Stored// 12 0028 0102 2 [2920 ] // IMG High Bit// 11 0028 0103 2 [2930 ] // IMG Pixel Representation// 0 0028 0106 2 [2940 ] // IMG Smallest Image Pixel Value// 0 0028 0107 2 [2950 ] // IMG Largest Image Pixel Value// 2601 0028 1050 4 [2960 ] // IMG Window Center//764 0028 1051 4 [2972 ] // IMG Window Width//1729 0028 1055 6 [2984 ] //IMG Window Center & Width Explanation//Algo1 Group: 0029, Length: 104190 0029 0010 18 [2998 ] // //SIEMENS CSA HEADER 0029 0011 22 [3024 ] // //SIEMENS MEDCOM HEADER2 0029 1008 12 [3054 ] // //IMAGE NUM 4 0029 1009 8 [3074 ] // //20130110 0029 1010 11524 [3094 ] // //hex OB: (len 8)53 56 31 30 04 03 02 01 0029 1018 2 [14626 ] // //MR 0029 1019 8 [14636 ] // //20130110 0029 1020 92520 [14656 ] // //hex OB: (len 8)53 56 31 30 04 03 02 01 0029 1160 4 [107184 ] // //com Group: 0032, Length: 18 0032 1060 10 [107196 ] //SDY Requested Procedure Description//TRIO HEAD Group: 0040, Length: 80 0040 0244 8 [107214 ] // PRC PPS Start Date//20130110 0040 0245 14 [107230 ] // PRC PPS Start Time//115038.781000 0040 0253 16 [107252 ] // PRC PPS ID//MR20130110115038 0040 0254 10 [107276 ] // PRC PPS Description//TRIO^HEAD Group: 0051, Length: 256 0051 0010 18 [107294 ] // //SIEMENS MR HEADER 0051 1008 12 [107320 ] // //IMAGE NUM 4 0051 1009 4 [107340 ] // //1.0 0051 100a 8 [107352 ] // //TA 00.01 0051 100b 6 [107368 ] // //64*64 0051 100c 14 [107382 ] // //FoV 1344*1344 0051 100e 24 [107404 ] // //Tra>Cor(-10.5)>Sag(1.1) 0051 100f 10 [107436 ] // //t:HEA;HEP 0051 1011 2 [107454 ] // //p2 0051 1012 4 [107464 ] // //TP 0 0051 1013 4 [107476 ] // //+LPH 0051 1015 4 [107488 ] // //D/R 0051 1016 14 [107500 ] // //p2 M/ND/MOSAIC 0051 1017 6 [107522 ] // //SL 3.0 0051 1019 6 [107536 ] // //A1/FS Group: 7fe0, Length: 401416 7fe0 0010 401408 [107554 ] // PXL Pixel Data//Data on disk DCM Dump Elements Complete Pixel array offset = 107554 (bytes) Pixel array length = 401408 (bytes)

frun_stimBOLD_mechanical_01, fslhd putput `cn2107:~ moana004$ fslhd '/home/moanae/moana004/project_FM_Gracely/BIDS_output/sub-HC0003/func/sub-HC0003_task-mechanical_run-02_bold.nii.gz' filename /home/moanae/moana004/project_FM_Gracely/BIDS_output/sub-HC0003/func/sub-HC0003_task-mechanical_run-02_bold.nii.gz

sizeof_hdr 348 data_type INT16 dim0 4 dim1 64 dim2 64 dim3 49 dim4 208 dim5 1 dim6 1 dim7 1 vox_units mm time_units s datatype 4 nbyper 2 bitpix 16 pixdim0 -1.000000 pixdim1 3.000000 pixdim2 3.000000 pixdim3 3.000000 pixdim4 3.000000 pixdim5 0.000000 pixdim6 0.000000 pixdim7 0.000000 vox_offset 352 cal_max 0.000000 cal_min 0.000000 scl_slope 1.000000 scl_inter 0.000000 phase_dim 2 freq_dim 1 slice_dim 3 slice_name sequential_increasing slice_code 1 slice_start 0 slice_end 0 slice_duration 0.000000 toffset 0.000000 intent Unknown intent_code 0 intent_name intent_p1 0.000000 intent_p2 0.000000 intent_p3 0.000000 qform_name Scanner Anat qform_code 1 qto_xyz:1 -2.997638 0.104681 0.056639 88.630219 qto_xyz:2 0.092629 2.948310 -0.546707 -75.112961 qto_xyz:3 0.074739 0.544528 2.949221 -56.226753 qto_xyz:4 0.000000 0.000000 0.000000 1.000000 qform_xorient Right-to-Left qform_yorient Posterior-to-Anterior qform_zorient Inferior-to-Superior sform_name Scanner Anat sform_code 1 sto_xyz:1 -2.997638 0.104680 0.056628 88.630219 sto_xyz:2 0.092629 2.948310 -0.546707 -75.112961 sto_xyz:3 0.074729 0.544528 2.949221 -56.226753 sto_xyz:4 0.000000 0.000000 0.000000 1.000000 sform_xorient Right-to-Left sform_yorient Posterior-to-Anterior sform_zorient Inferior-to-Superior file_type NIFTI-1+ file_code 1 descrip TE=30;Time=123528.355;phase=1 aux_file
`

[mharms]

Attempting to set a "P>A" polarity using Phase enc. dir. = P>>A is prone to a Siemens bug that will actually flip the polarity to R/L (or L/R, I don't recall which). What do the sidecar json files report for the "Phase encoding direction"? The difference should also be obvious if you just look at the direction of the distortions in the images.

[estephanmoana]

Indeed, I noticed the issue when doing QC of the preprocessed imaged using fmriprep. I copied the content of the json files for the two BOLD runs I mentioned. How do you recommend me to proceed? Thank you.

sub-HC0003_task-auditory_run-03_bold.json

{ "AcquisitionMatrixPE": 64, "AcquisitionNumber": 1, "AcquisitionTime": "12:52:32.330000", "BandwidthPerPixelPhaseEncode": 60.096, "BaseResolution": 64, "CoilCombinationMethod": "Adaptive Combine", "CoilString": "t:HEA;HEP", "ConversionSoftware": "dcm2niix", "ConversionSoftwareVersion": "v1.0.20211006", "DerivedVendorReportedEchoSpacing": 0.000520001, "DeviceSerialNumber": "35412", "DwellTime": 3.4e-06, "EchoTime": 0.03, "EffectiveEchoSpacing": 0.000260001, "FlipAngle": 77, "HeudiconvVersion": "0.11.3+d20220512", "ImageOrientationPatientDICOM": [ 0.999213, -0.0308765, 0.0249095, 0.0348933, 0.98277, -0.181509 ], "ImageOrientationText": "Tra>Cor(-10.5)>Sag(1.1)", "ImageType": [ "ORIGINAL", "PRIMARY", "M", "ND", "MOSAIC" ], "ImagingFrequency": 123.196, "InPlanePhaseEncodingDirectionDICOM": "ROW", "InstitutionAddress": "Mason Farm Rd 106,Chapel Hill,District,US,27599", "InstitutionName": "University of North Carolina", "InstitutionalDepartmentName": "Department", "MRAcquisitionType": "2D", "MagneticFieldStrength": 3, "Manufacturer": "Siemens", "ManufacturersModelName": "TrioTim", "MatrixCoilMode": "GRAPPA", "Modality": "MR", "ParallelReductionFactorInPlane": 2, "PartialFourier": 1, "PatientPosition": "HFS", "PercentPhaseFOV": 100, "PercentSampling": 100, "PhaseEncodingDirection": "i", "PhaseEncodingSteps": 63, "PhaseResolution": 1, "PixelBandwidth": 2298, "ProcedureStepDescription": "TRIO^HEAD", "ProtocolName": "BOLD auditory 3x3x3", "PulseSequenceDetails": "%SiemensSeq%\\ep2d_bold", "ReceiveCoilName": "HeadMatrix", "ReconMatrixPE": 64, "RefLinesPE": 24, "RepetitionTime": 3, "SAR": 0.113101, "ScanOptions": "FS", "ScanningSequence": "EP", "SequenceName": "*epfid2d1_64", "SequenceVariant": "SK", "SeriesDescription": "BOLD auditory 3x3x3", "SeriesNumber": 14, "ShimSetting": [ -3263, 8397, 13679, -208, -549, -1310, 51, 1362 ], "SliceThickness": 3, "SliceTiming": [ 0, 0.0625, 0.125, 0.185, 0.2475, 0.3075, 0.37, 0.4325, 0.4925, 0.555, 0.615, 0.6775, 0.7375, 0.8, 0.8625, 0.9225, 0.985, 1.045, 1.1075, 1.17, 1.23, 1.2925, 1.3525, 1.415, 1.475, 1.5375, 1.6, 1.66, 1.7225, 1.7825, 1.845, 1.9075, 1.9675, 2.03, 2.09, 2.1525, 2.2125, 2.275, 2.3375, 2.3975, 2.46, 2.52, 2.5825, 2.645, 2.705, 2.7675, 2.8275, 2.89, 2.95 ], "SoftwareVersions": "syngo MR B17", "SpacingBetweenSlices": 3, "StationName": "MEDPC", "TaskName": "auditory", "TotalReadoutTime": 0.01638, "TxRefAmp": 290.174, "NumberOfVolumesDiscardedByScanner": 3, "NumberOfVolumesDiscardedByUser": 0 }

sub-HC0003_task-mechanical_run-02_bold.json

{ "AcquisitionMatrixPE": 64, "AcquisitionNumber": 1, "AcquisitionTime": "12:35:28.355000", "BandwidthPerPixelPhaseEncode": 60.096, "BaseResolution": 64, "CoilCombinationMethod": "Adaptive Combine", "CoilString": "t:HEA;HEP", "ConversionSoftware": "dcm2niix", "ConversionSoftwareVersion": "v1.0.20211006", "DerivedVendorReportedEchoSpacing": 0.000520001, "DeviceSerialNumber": "35412", "DwellTime": 3.4e-06, "EchoTime": 0.03, "EffectiveEchoSpacing": 0.000260001, "FlipAngle": 77, "HeudiconvVersion": "0.11.3+d20220512", "ImageOrientationPatientDICOM": [ 0.999213, -0.0308765, 0.0249095, 0.0348933, 0.98277, -0.181509 ], "ImageOrientationText": "Tra>Cor(-10.5)>Sag(1.1)", "ImageType": [ "ORIGINAL", "PRIMARY", "M", "ND", "MOSAIC" ], "ImagingFrequency": 123.196, "InPlanePhaseEncodingDirectionDICOM": "COL", "InstitutionAddress": "Mason Farm Rd 106,Chapel Hill,District,US,27599", "InstitutionName": "University of North Carolina", "InstitutionalDepartmentName": "Department", "MRAcquisitionType": "2D", "MagneticFieldStrength": 3, "Manufacturer": "Siemens", "ManufacturersModelName": "TrioTim", "MatrixCoilMode": "GRAPPA", "Modality": "MR", "ParallelReductionFactorInPlane": 2, "PartialFourier": 1, "PatientPosition": "HFS", "PercentPhaseFOV": 100, "PercentSampling": 100, "PhaseEncodingDirection": "j-", "PhaseEncodingSteps": 63, "PhaseResolution": 1, "PixelBandwidth": 2298, "ProcedureStepDescription": "TRIO^HEAD", "ProtocolName": "BOLD mechanical 3x3x3", "PulseSequenceDetails": "%SiemensSeq%\\ep2d_bold", "ReceiveCoilName": "HeadMatrix", "ReconMatrixPE": 64, "RefLinesPE": 24, "RepetitionTime": 3, "SAR": 0.113101, "ScanOptions": "FS", "ScanningSequence": "EP", "SequenceName": "*epfid2d1_64", "SequenceVariant": "SK", "SeriesDescription": "BOLD mechanical 3x3x3", "SeriesNumber": 12, "ShimSetting": [ -3263, 8397, 13679, -208, -549, -1310, 51, 1362 ], "SliceThickness": 3, "SliceTiming": [ 0, 0.06, 0.1225, 0.1825, 0.245, 0.3075, 0.3675, 0.43, 0.49, 0.5525, 0.6125, 0.675, 0.7375, 0.7975, 0.86, 0.92, 0.9825, 1.045, 1.105, 1.1675, 1.2275, 1.29, 1.35, 1.4125, 1.475, 1.535, 1.5975, 1.6575, 1.72, 1.7825, 1.8425, 1.905, 1.965, 2.0275, 2.0875, 2.15, 2.2125, 2.2725, 2.335, 2.395, 2.4575, 2.52, 2.58, 2.6425, 2.7025, 2.765, 2.825, 2.8875, 2.95 ], "SoftwareVersions": "syngo MR B17", "SpacingBetweenSlices": 3, "StationName": "MEDPC", "TaskName": "mechanical", "TotalReadoutTime": 0.01638, "TxRefAmp": 290.174, "NumberOfVolumesDiscardedByScanner": 3, "NumberOfVolumesDiscardedByUser": 0 }

[mharms]

There isn't much you can in terms of distortion correction unless the tool you are using happens to support the situation in which the BOLD scan has a different PE axis from the SE EPI scans that you (presumably) using with topup for the correction.

[pvelasco]

I agree with @mharms that often, if you are not careful, the scanner would switch the PE direction without you noticing. I would trust the results of dcm2niix (extracted from the DICOM header) over some sequence printout, which would correspond to a saved protocol, not necessarily to the acquired images. In principle, as long as topup estimates the correct field, applytopup should be able to apply the distortion correction to an acquisition with distortions along the other direction, since at that point the correction should be independent of the method you used to measure/estimate the field. The only point that could cause trouble is when trying to correct for motion between the SE EPI scans and the BOLD scan. But, if applytopup is smart enough, it would try to estimate relative motion in the undistorted space. So, again, not depending on the SE EPI and BOLD scans having the same distortion directions. I think it is worth giving it a try, and if you get images that are not properly unwarped, you can ask at the FSL mailing list for advice.

[mharms]

Yes, in principle I think direct use of topup/applytopup should be able to handle this situation. The problem is that other integrated pipelines may not support this flexibility. e.g., the HCPpipelines currently have a hard-coded assumption that the SEFMs and BOLD scans have the same PE axis. I'm not sure if fmriprep makes a similar assumption.

[neurolabusc]

@estephanmoana dcm2niix faithfully reports the DICOM data as acquired, not as planned. Looking at your PDFs, the intention was to acquire phase reversed sequences A>>P and P>>A by using the Rotation of 180 degrees. As @mharms and @pvelasco have noted, a bug in the Siemens product sequence caused the requested 180 degree rotation to be applied as a 90 degree (RL) rotation.

The rationale of TOPUP is a A>>P and P>>A scan have spatial distortion of equal magnitude but opposite direction, so you can compute a nonlinear transform that splits the difference between these two images to infer the undistorted shape. While TOPUP allows some flexibility in modeling the distortion, I think having an A>>P and R>>L scan would be unlikely to yield a nice result.

For future acquisitions, you have two options to avoid this bug:

1. Work with you Siemens Research Collaboration Manager to get software patch that resolves this. I am not sure what the patch was for your VB17 system, for our VE11C system the critical patch was Service Help 05 (SH05) which was released on December 20, 2018. Be aware that this patch did enforce more stringent stimulation limits to comply with IEC 60601-2-33, so when you get it applied run all your sequences on a phantom to see all are acquired successfully or whether they need to be adjusted.
2. The fantastic CMMR research sequences include a check box in the Special tab that allows you to reverse readout/phase encoding polarity. This works regardless of the Siemens software version.

[mharms]

@neurolabusc My understanding was that this Siemens PE bug in their product EPI sequences never got fixed in VE11, and in fact may still be an issue in XA30

[estephanmoana]

Thanks for all this great information everyone, very helpful! I did not hear about this bug in the Siemens software until now, it just adds another layer of information to be aware of when processing MRI data, not a simple task by any means :) For my current and future studies, I am using the publicly available HCP imaging sequences, but this is a legacy data set acquired between 2011-2013 so I am trying to use the best methods possible for it within its limitations.

So it seems that I cannot used the GRE field maps for the BOLD runs acquired in this data set for distortion correction. I am considering using the fieldmap-free distortion correction option in fmriprep that uses. See description from Esteban et al., 2019 below. What are your thoughts on this, is this better than no distortion correction at all? Thank you.

Fieldmap-less susceptibility distortion correction. Many legacy and current human fMRI protocols lack the magnetic resonance fieldmaps necessary to apply standard SDC methods. As described in Supplementary Fig. 6, the BIDS dataset is queried to discover whether extra acquisitions containing fieldmap information are available. When no fieldmap information is found, fMRIPrep adapts the fieldmap-less correction for diffusion EPI images introduced by Wang et al.105. They propose using the same-subject T1w reference as the undistorted target in a nonlinear registration scheme. To maximize the similarity between the T2* contrast of the EPI scan and the reference T1w data, the intensities of the latter are inverted. To regularize the optimization of the deformation field, only displacements along the phase-encoding direction are allowed, and the magnitude of the displacements is modulated using priors. To our knowledge, no other existing pipeline applies fieldmap-less SDC to BOLD images. Further details on the integration of the different SDC techniques, and particularly this fieldmap-less option, are presented in Supplementary Note 3.

From Supplementary Note 3.

Fieldmap-less susceptibility distortion correction (SDC). This workflow takes a skull-stripped T1w image and reference BOLD image, and estimates a field of displacements that compensates for the warp caused by susceptibility distortion. The tool uses ANTs’ antsRegistration configured with symmetric normalization (SyN) to align a fieldmap template13 and applies the template as prior information to regularize a follow-up registration process. The follow-up registration process also uses antsRegistration with SyN deformation, with displacements restricted to the phase-encoding (PE) direction. If no PE direction is specified, anterior-posterior PE is assumed. Based on the fieldmap atlas, the displacement field is optimized only within regions that are expected to have a >3mm (approximately 1 voxel) warp. This technique is a variation on previous work11,12.

[mharms]

I would try to figure out how to use the GRE (presumably dual echo?) field maps that you have available. It should be possible, but you'll need to dig into and understand the tools that you using.

[captainnova]

Siemens did fix the bug in XA. (Verified with xa20, IIRC.) Is CMRR's WIP available for XA now?

Correcting epi distortion with a pa and an rl is possible, sort of, but you will need to get the timing parameters correct, and the last time I tried (years ago), it had a tendency to introduce diagonal stripes.

Rob

On Mon, Sep 5, 2022, 4:50 PM Michael Harms @.***> wrote:

@neurolabusc https://github.com/neurolabusc My understanding was that this Siemens PE bug in their product EPI sequences never got fixed in VE11, and in fact may still be an issue in XA30

— Reply to this email directly, view it on GitHub https://github.com/rordenlab/dcm2niix/issues/632#issuecomment-1237481339, or unsubscribe https://github.com/notifications/unsubscribe-auth/ADY6J2O3YAJ3PIEX4VE3G33V4ZTJ5ANCNFSM6AAAAAAQFERMMA . You are receiving this because you are subscribed to this thread.Message ID: @.***>

[mharms]

Interesting, because in talking to some others, I'm getting the sense that Siemens is looking to some "addin" as a solution to the PE polarity bug, which would seem to imply that it isn't entirely fixed in XA30.

[estephanmoana]

As I'm trying to better understand what data were actually acquired (instead of what is described the scanner printout sequence - see attached), I extracted the phase encoding direction from the json files for all participants in the study. In the example participant below, one can see that a few files have PE "i" while the majority had "j-", but no file has PE "j".

In the scanner printout, the PE for field maps is "A >>P" while the PE of the EPI images that are to be corrected is "P >> A". My question then is: shouldn't the field maps have PE = "j-" and the EPI images have PE = "j"?

subjID	File name	Data type	File type	Image contrast	PE DICOM	PE json
sub-FM0001	sub-FM0001_acq-asl_run-01_magnitude1.json	fmap	magnitude1	asl	COL	j-
sub-FM0001	sub-FM0001_acq-asl_run-01_magnitude2.json	fmap	magnitude2	asl	COL	j-
sub-FM0001	sub-FM0001_acq-asl_run-01_phasediff.json	fmap	phasediff	asl	COL	j-
sub-FM0001	sub-FM0001_run-01_asl.json	perf	asl	asl	COL	j-
sub-FM0001	sub-FM0001_run-02_asl.json	perf	asl	asl	COL	j-
sub-FM0001	sub-FM0001_run-01_dwi.json	dwi	dwi	dwi	COL	j-
sub-FM0001	sub-FM0001_acq-dwi_run-01_magnitude1.json	fmap	magnitude1	dwi	COL	j-
sub-FM0001	sub-FM0001_acq-dwi_run-01_magnitude2.json	fmap	magnitude2	dwi	COL	j-
sub-FM0001	sub-FM0001_acq-dwi_run-01_phasediff.json	fmap	phasediff	dwi	COL	j-
sub-FM0001	sub-FM0001_acq-funcrest_run-01_magnitude1.json	fmap	magnitude1	task-rest	COL	j-
sub-FM0001	sub-FM0001_acq-funcrest_run-01_magnitude2.json	fmap	magnitude2	task-rest	COL	j-
sub-FM0001	sub-FM0001_acq-funcrest_run-01_phasediff.json	fmap	phasediff	task-rest	COL	j-
sub-FM0001	sub-FM0001_task-rest_run-01_bold.json	func	bold-rest	task-rest	ROW	i
sub-FM0001	sub-FM0001_task-rest_run-02_bold.json	func	bold-rest	task-rest	COL	j-
sub-FM0001	sub-FM0001_acq-funcstim_run-01_magnitude1.json	fmap	magnitude1	task-stim	COL	j-
sub-FM0001	sub-FM0001_acq-funcstim_run-01_magnitude2.json	fmap	magnitude2	task-stim	COL	j-
sub-FM0001	sub-FM0001_acq-funcstim_run-01_phasediff.json	fmap	phasediff	task-stim	COL	j-
sub-FM0001	sub-FM0001_task-auditory_run-03_bold.json	func	bold-stim	task-stim-auditory	ROW	i
sub-FM0001	sub-FM0001_task-intermingled_run-01_bold.json	func	bold-stim	task-stim-intermingled	ROW	i
sub-FM0001	sub-FM0001_task-mechanical_run-02_bold.json	func	bold-stim	task-stim-mechanical	COL	j-
sub-FM0001	sub-FM0001_task-simultaneous_run-04_bold.json	func	bold-stim	task-stim-simultaneous	COL	j-

Gracely_MR pulse sequence_20111119.pdf

[pvelasco]

I see. So you don't have any SE EPI field maps, but only GRE field maps.

In that case, it doesn't matter what PE those had, since they are virtually undistorted (they are not EPI sequences).

What tool are you trying to use to un-distort the functional images? It's been a very long time since I have used GRE images to unwarp EPIs, but FSL's fugue did the job. I don't know for sure if fmriprep would handle those, but I think it might...

[mharms]

If you happen to be using the HCPpipelines, or those within the context of QuNex, use of dual-echo GRE field maps is supported.

[estephanmoana]

My first attempt at processing these data was using QuNex using "LegacyData" mode (actually I spent a 1 1/2 month learning to use QuNex with these data), however I got stuck at the MSMAll and DeDriftAndResample steps, since the scripts are not designed to handle "LegacyData" mode gracefully (the lack of myelin maps is not acconted for in those scripts). I tried to manually change these scripts to avoid the issue of no myelin maps but it became overwhelming, that's when I decided to give fmriprep a try.

Please see below the SDC results for the example participant I mentioned in the previous message. Some distortion correction makes things worse, while others seem reasonable. So I guess that fmriprep is not able to fully account for the different PE in some of the BOLD runs for some reason.

fMRIPrep version: 22.0.0 fMRIPrep command: /opt/conda/bin/fmriprep /home/moanae/moana004/project_FM_Gracely/BIDS_output /home/moanae/shared/project_FM_Gracely_process01/anat-func_preprocess participant --participant-label sub-FM0001 --n-cpus 8 --omp-nthreads 8 --error-on-aroma-warnings --skip_bids_validation --ignore slicetiming --output-spaces MNI152NLin2009cAsym:res-2 MNI152NLin6Asym:res-2 fsLR --bold2t1w-init register --bold2t1w-dof 12 --skull-strip-t1w force --use-aroma --cifti-output 91k --debug fieldmaps --work-dir /tmp/tempfiles_fmriprep --fs-license-file /panfs/roc/msisoft/freesurfer/6.0.0/license.txt --resource-monitor --stop-on-first-crash

[estephanmoana]

Link to the html report https://umn.box.com/s/jn4fa58ar1jb26n423stkwmpfz2v91a9

[mharms]

The HCPpipelines support the use of dual-echo GRE fieldmaps for SDC. You should not have needed to use LegacyData mode for basic fMRIVolume processing.

Is fmriprep equipped to make use of dual-echo GRE fieldmaps? If not, it is almost certainly doing something nonsensical.

[mharms]

At this point, your issues have nothing to do with dcm2niix. I suggest you take your inquiries to the other appropriate support groups.

[estephanmoana]

Sure, I look into this in other forums. I appreciate all the information here, thanks to all.