Re: [Freesurfer] convert FS volume in scanner space
Thanks for getting back. What I need is hippoSflabels in the original coordinates/space of the dicom-files (to compare with manual segmentation). Basically I had success with mri_convert -rl rawavg.mgz -rt nearest lh.hippoSfLabels-T1.v10.mgz FS_outfile.nii (identical dimensions, voxel spacing, origin and orientation) however, segmentation have wholes that renders shape analysis impossible. Running mri_vol2vol --mov lh.hippoSfLabels-T1.v10.mgz --targ orig/001.mgz --regheader --o FS_outfile.nii --no-save-reg (or —targ rawavg) resulted in perfect segmentations, but the coordinates had (identical dimensions and voxel spacing) but different origin and orientation (dicom: AIL; FS_outfile: ASR) when I open both files in ITK Snap. Also, FS seems to involve more/ other steps in mri_vol2vol as compared to mri_convert. So I figured I would either need other commands for mri_vol2vol or an additional step to have my (perfect) segmention in dicom-coordinates. I tried mri_convert FS_outfile.nii FS_outfile_LPS.nii --in_orientation RAS --out_orientation LPS but the output was completely wrong oriented. So that’s were I am stuck. Back to your question - I guess I need the voxel-to-RAS matrix to have FS-output in the original real world coordinates to enable direct comparisons and then guidance on how to calculate the image translations with FS (on a larger data set). Am 09.07.2015 um 19:00 schrieb Douglas N Greve gr...@nmr.mgh.harvard.edumailto:gr...@nmr.mgh.harvard.edu: There are a lot of details here that can easily get confused. The pixel data are not in any particular space. The order of the pixel data are not changed when converted to mgz. The direction cosines that indicate how to convert a col-row-slice to a real world coordinate dictate how the pixels are displayed or interpreted spatially. The direction cosines in the dicom file will be LPS. The direction cosines in RAS space are computed and saved in the mgz file. The coordinate center will not change. Thus if you use the voxel-to-RAS matrix to compute the location in the scanner of a given col-row-slice, it will do so accurately. Given that, can you explain more about what you actually need? On 07/08/2015 10:51 AM, von Polier, Georg wrote: Sorry for the confusion: According to the presentation fswiki/CoordinateSystems (below) I meant, the data are in native space/ RAS, with coordinate center not at magnet center, which is what I need. Scanner Space - coordinate center at magnet isocenter, bore axis is Z, X to the left, Y to the ceiling. Direction cosines and P0 defined in DICOM file (note: this is an LPS, not RAS coordinate system) Native - basically the same as scanner, but RAS. -Ursprüngliche Nachricht- Von: freesurfer-boun...@nmr.mgh.harvard.edumailto:freesurfer-boun...@nmr.mgh.harvard.edu [mailto:freesurfer-boun...@nmr.mgh.harvard.edu] Im Auftrag von Douglas Greve Gesendet: Mittwoch, 8. Juli 2015 16:37 An: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Betreff: Re: [Freesurfer] convert FS volume in scanner space what do you mean the coords are still in RAS? On 7/8/15 8:25 AM, von Polier, Georg wrote: My command line was cd $SUBJECTS_DIR/subjid/mri mri_vol2vol --mov brain.mgz --targ rawavg.mgz --regheader --o brain-in-rawavg.mgz --no-save-reg according to fswiki/FsAnat-to-NativeAnat. As I mentioned, I am happy with the results but the coordinates are still RAS (from my understanding) and different from the original dicom-coordinates (that my manual segmentations are in). When I use mri_convert, the coordinates are in dicom-space (0 in center of image/ coil; offset and z-axis identical with original dicoms), but the segmentation seems different/ more prone to artifacts, also when I use rt- nearest. Georg -Ursprüngliche Nachricht- Von: freesurfer-boun...@nmr.mgh.harvard.edumailto:freesurfer-boun...@nmr.mgh.harvard.edu [mailto:freesurfer-boun...@nmr.mgh.harvard.edu] Im Auftrag von Douglas Greve Gesendet: Mittwoch, 8. Juli 2015 04:20 An: freesurfer@nmr.mgh.harvard.edumailto:freesurfer@nmr.mgh.harvard.edu Betreff: Re: [Freesurfer] convert FS volume in scanner space what was your vol2vol command line? What wiki page were you referencing? What was the problem with the coordinates? On 7/7/15 8:11 PM, Bruce Fischl wrote: I think you meant you want nearest neighbor interpolation, not trilinear, since you are mapping labels. Try using -rt nearest insted On Wed, 8 Jul 2015, von Polier, Georg wrote: Thanks for your reply. I tried mri_convert with -rt interpolate (resulting in trilinear interpolation), however, the results were partially prone to artifacts and different from those I get with mir_vol2vol (but in dicom space). Is there a way using mir_vol2vol with results in dicom space, or a second step that gives a vox2vox from conformed/anatomical space to dicom space? I use the .mgz-output of hippo_subfields (FS_dev). Thanks again, Georg Am 07.07.2015 um 02:34
Re: [Freesurfer] convert FS volume in scanner space
Different in nii: type: FLOAT (3); TE, TI, flip angle: 0; PhEncDir: Unknown; xform info, voxel to ras transform, ras to voxel identical Loading rawavg.nii and FS_outfile.nii in ITK looks in perfect alignment, it says orientation of FS_outfile is Oblique (closest to ASR); When I load the manual segents, (into rawavg) it warns of header mismatch (orientation ASL) and I notice a slight shift. Loading the FS_outfile into the original dicom-file, it equally warns of header missmatch, I notice a slight shift. Volume information for rawavg.mgz type: MGH dimensions: 256 x 256 x 176 voxel sizes: 0.9766, 0.9766, 1. type: SHORT (4) fov: 250.000 dof: 0 xstart: -125.0, xend: 125.0 ystart: -125.0, yend: 125.0 zstart: -88.0, zend: 88.0 TR: 1900.00 msec, TE: 2.52 msec, TI: 900.00 msec, flip angle: 9.00 degrees nframes: 1 PhEncDir: ROW ras xform present xform info: x_r = -0.0332, y_r = -0.0628, z_r = -0.9975, c_r =-3.9057 : x_a = -0.9995, y_a = 0.0021, z_a = 0.0331, c_a =14.5730 : x_s = 0., y_s = -0.9980, z_s = 0.0628, c_s = -24.5551 talairach xfm : Orientation : PIL Primary Slice Direction: sagittal voxel to ras transform: -0.0324 -0.0613 -0.997595.8613 -0.9760 0.0020 0.0331 136.3322 0. -0.9746 0.062894.6727 0. 0. 0. 1. voxel-to-ras determinant -0.953674 ras to voxel transform: -0.0340 -1.0234 0. 142.7820 -0.0643 0.0021 -1.0220 102.6232 -0.9975 0.0331 0.062885.1638 0. 0. 0. 1. Volume information for FS_output.nii type: nii dimensions: 256 x 256 x 176 voxel sizes: 0.9766, 0.9766, 1. type: FLOAT (3) fov: 250.000 dof: 0 xstart: -125.0, xend: 125.0 ystart: -125.0, yend: 125.0 zstart: -88.0, zend: 88.0 TR: 1900.00 msec, TE: 0.00 msec, TI: 0.00 msec, flip angle: 0.00 degrees nframes: 1 PhEncDir: UNKNOWN ras xform present xform info: x_r = -0.0332, y_r = -0.0628, z_r = -0.9975, c_r =-3.9057 : x_a = -0.9995, y_a = 0.0021, z_a = 0.0331, c_a =14.5730 : x_s = 0., y_s = -0.9980, z_s = 0.0628, c_s = -24.5551 Orientation : PIL Primary Slice Direction: sagittal voxel to ras transform: -0.0324 -0.0613 -0.997595.8613 -0.9760 0.0020 0.0331 136.3322 0. -0.9746 0.062894.6727 0. 0. 0. 1. voxel-to-ras determinant -0.953674 ras to voxel transform: -0.0340 -1.0234 0. 142.7820 -0.0643 0.0021 -1.0220 102.6232 -0.9975 0.0331 0.062885.1638 0. 0. 0. 1. Am 09.07.2015 um 20:54 schrieb Douglas N Greve gr...@nmr.mgh.harvard.edumailto:gr...@nmr.mgh.harvard.edu: vol2vol is the right way to go on this. When you run mri_info on rawavg.mgz and FS_outfile.nii, do they have the same geometry? What if you convert rawavg.mgz to nii and open it and outfile.nii in ITK Snap, so they look the same? On 07/09/2015 02:44 PM, von Polier, Georg wrote: Thanks for getting back. What I need is hippoSflabels in the original coordinates/space of the dicom-files (to compare with manual segmentation). Basically I had success with mri_convert -rl rawavg.mgz -rt nearest lh.hippoSfLabels-T1.v10.mgz FS_outfile.nii (identical dimensions, voxel spacing, origin and orientation) however, segmentation have wholes that renders shape analysis impossible. Running mri_vol2vol --mov lh.hippoSfLabels-T1.v10.mgz --targ orig/001.mgz --regheader --o FS_outfile.nii --no-save-reg (or —targ rawavg) resulted in perfect segmentations, but the coordinates had (identical dimensions and voxel spacing) but different origin and orientation (dicom: AIL; FS_outfile: ASR) when I open both files in ITK Snap. Also, FS seems to involve more/ other steps in mri_vol2vol as compared to mri_convert. So I figured I would either need other commands for mri_vol2vol or an additional step to have my (perfect) segmention in dicom-coordinates. I tried mri_convert FS_outfile.nii FS_outfile_LPS.nii --in_orientation RAS --out_orientation LPS but the output was completely wrong oriented. So that’s were I am stuck. Back to your question - I guess I need the voxel-to-RAS matrix to have FS-output in the original real world coordinates to enable direct comparisons and then guidance on how to calculate the image translations with FS (on a larger data set). Am 09.07.2015 um 19:00 schrieb Douglas N Greve gr...@nmr.mgh.harvard.edumailto:gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu: There are a lot
Re: [Freesurfer] convert FS volume in scanner space
The vol that holds the man segs was created by ITK Snap; your command line worked for me, THANKS SO MUCH!! Georg Am 09.07.2015 um 22:12 schrieb Douglas N Greve gr...@nmr.mgh.harvard.edu: How was the volume that holds the manual segs created? Another thing you can try is mri_vol2vol --mov lh.hippoSfLabels-T1.v10.mgz --targ mansegs.nii --regheader --o FS_outfile.nii On 07/09/2015 03:59 PM, von Polier, Georg wrote: Different in nii: type: FLOAT (3); TE, TI, flip angle: 0; PhEncDir: Unknown; xform info, voxel to ras transform, ras to voxel identical Loading rawavg.nii and FS_outfile.nii in ITK looks in perfect alignment, it says orientation of FS_outfile is Oblique (closest to ASR); When I load the manual segents, (into rawavg) it warns of header mismatch (orientation ASL) and I notice a slight shift. Loading the FS_outfile into the original dicom-file, it equally warns of header missmatch, I notice a slight shift. Volume information for rawavg.mgz type: MGH dimensions: 256 x 256 x 176 voxel sizes: 0.9766, 0.9766, 1. type: SHORT (4) fov: 250.000 dof: 0 xstart: -125.0, xend: 125.0 ystart: -125.0, yend: 125.0 zstart: -88.0, zend: 88.0 TR: 1900.00 msec, TE: 2.52 msec, TI: 900.00 msec, flip angle: 9.00 degrees nframes: 1 PhEncDir: ROW ras xform present xform info: x_r = -0.0332, y_r = -0.0628, z_r = -0.9975, c_r = -3.9057 : x_a = -0.9995, y_a = 0.0021, z_a = 0.0331, c_a = 14.5730 : x_s = 0., y_s = -0.9980, z_s = 0.0628, c_s = -24.5551 talairach xfm : Orientation : PIL Primary Slice Direction: sagittal voxel to ras transform: -0.0324 -0.0613 -0.997595.8613 -0.9760 0.0020 0.0331 136.3322 0. -0.9746 0.062894.6727 0. 0. 0. 1. voxel-to-ras determinant -0.953674 ras to voxel transform: -0.0340 -1.0234 0. 142.7820 -0.0643 0.0021 -1.0220 102.6232 -0.9975 0.0331 0.062885.1638 0. 0. 0. 1. Volume information for FS_output.nii type: nii dimensions: 256 x 256 x 176 voxel sizes: 0.9766, 0.9766, 1. type: FLOAT (3) fov: 250.000 dof: 0 xstart: -125.0, xend: 125.0 ystart: -125.0, yend: 125.0 zstart: -88.0, zend: 88.0 TR: 1900.00 msec, TE: 0.00 msec, TI: 0.00 msec, flip angle: 0.00 degrees nframes: 1 PhEncDir: UNKNOWN ras xform present xform info: x_r = -0.0332, y_r = -0.0628, z_r = -0.9975, c_r = -3.9057 : x_a = -0.9995, y_a = 0.0021, z_a = 0.0331, c_a = 14.5730 : x_s = 0., y_s = -0.9980, z_s = 0.0628, c_s = -24.5551 Orientation : PIL Primary Slice Direction: sagittal voxel to ras transform: -0.0324 -0.0613 -0.997595.8613 -0.9760 0.0020 0.0331 136.3322 0. -0.9746 0.062894.6727 0. 0. 0. 1. voxel-to-ras determinant -0.953674 ras to voxel transform: -0.0340 -1.0234 0. 142.7820 -0.0643 0.0021 -1.0220 102.6232 -0.9975 0.0331 0.062885.1638 0. 0. 0. 1. Am 09.07.2015 um 20:54 schrieb Douglas N Greve gr...@nmr.mgh.harvard.edu mailto:gr...@nmr.mgh.harvard.edu: vol2vol is the right way to go on this. When you run mri_info on rawavg.mgz and FS_outfile.nii, do they have the same geometry? What if you convert rawavg.mgz to nii and open it and outfile.nii in ITK Snap, so they look the same? On 07/09/2015 02:44 PM, von Polier, Georg wrote: Thanks for getting back. What I need is hippoSflabels in the original coordinates/space of the dicom-files (to compare with manual segmentation). Basically I had success with mri_convert -rl rawavg.mgz -rt nearest lh.hippoSfLabels-T1.v10.mgz FS_outfile.nii (identical dimensions, voxel spacing, origin and orientation) however, segmentation have wholes that renders shape analysis impossible. Running mri_vol2vol --mov lh.hippoSfLabels-T1.v10.mgz --targ orig/001.mgz --regheader --o FS_outfile.nii --no-save-reg (or —targ rawavg) resulted in perfect segmentations, but the coordinates had (identical dimensions and voxel spacing) but different origin and orientation (dicom: AIL; FS_outfile: ASR) when I open both files in ITK Snap. Also, FS seems to involve more/ other steps in mri_vol2vol as compared to mri_convert. So I figured I would either need other commands for mri_vol2vol or an additional step to have my (perfect) segmention in dicom-coordinates. I tried mri_convert FS_outfile.nii FS_outfile_LPS.nii --in_orientation RAS --out_orientation LPS but the output was completely wrong oriented. So that’s were I am stuck. Back to your question - I guess I need the voxel-to-RAS matrix to have FS-output in the original real world coordinates to enable direct comparisons and then guidance on how to calculate the image translations with FS (on a larger data set
Re: [Freesurfer] convert FS volume in scanner space
My command line was cd $SUBJECTS_DIR/subjid/mri mri_vol2vol --mov brain.mgz --targ rawavg.mgz --regheader --o brain-in-rawavg.mgz --no-save-reg according to fswiki/FsAnat-to-NativeAnat. As I mentioned, I am happy with the results but the coordinates are still RAS (from my understanding) and different from the original dicom-coordinates (that my manual segmentations are in). When I use mri_convert, the coordinates are in dicom-space (0 in center of image/ coil; offset and z-axis identical with original dicoms), but the segmentation seems different/ more prone to artifacts, also when I use rt- nearest. Georg -Ursprüngliche Nachricht- Von: freesurfer-boun...@nmr.mgh.harvard.edu [mailto:freesurfer-boun...@nmr.mgh.harvard.edu] Im Auftrag von Douglas Greve Gesendet: Mittwoch, 8. Juli 2015 04:20 An: freesurfer@nmr.mgh.harvard.edu Betreff: Re: [Freesurfer] convert FS volume in scanner space what was your vol2vol command line? What wiki page were you referencing? What was the problem with the coordinates? On 7/7/15 8:11 PM, Bruce Fischl wrote: I think you meant you want nearest neighbor interpolation, not trilinear, since you are mapping labels. Try using -rt nearest insted On Wed, 8 Jul 2015, von Polier, Georg wrote: Thanks for your reply. I tried mri_convert with -rt interpolate (resulting in trilinear interpolation), however, the results were partially prone to artifacts and different from those I get with mir_vol2vol (but in dicom space). Is there a way using mir_vol2vol with results in dicom space, or a second step that gives a vox2vox from conformed/anatomical space to dicom space? I use the .mgz-output of hippo_subfields (FS_dev). Thanks again, Georg Am 07.07.2015 um 02:34 schrieb Bruce Fischl fis...@nmr.mgh.harvard.edu: Hi Georg if you have a volume (like the 001.mgz) that is in scanner coordinates you can use the reslice like and resample type flags in mri_convert: mri_convert -rl 001.mgz -rt trilinear input.mgz output.mgz cheers Bruce On Mon, 6 Jul 2015, von Polier, Georg wrote: Dear FreeSurfers, I need to convert an FS-volume back in scanner space (to calculate overlap with manual segmentation). I went through the wiki-entry and converted to native space with mir_vol2vol, however the coordinates are still RAS and not identical with the original Scanner-coordinates. I tried mri_convert, however, the results were not as good as mri_vol2vol (apparently need trilinear interpolation, not available in mri_convert). Any help would be greatly appreciated, Cheers, Georg ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Re: [Freesurfer] convert FS volume in scanner space
Thanks for your reply. I tried mri_convert with -rt interpolate (resulting in trilinear interpolation), however, the results were partially prone to artifacts and different from those I get with mir_vol2vol (but in dicom space). Is there a way using mir_vol2vol with results in dicom space, or a second step that gives a vox2vox from conformed/anatomical space to dicom space? I use the .mgz-output of hippo_subfields (FS_dev). Thanks again, Georg Am 07.07.2015 um 02:34 schrieb Bruce Fischl fis...@nmr.mgh.harvard.edu: Hi Georg if you have a volume (like the 001.mgz) that is in scanner coordinates you can use the reslice like and resample type flags in mri_convert: mri_convert -rl 001.mgz -rt trilinear input.mgz output.mgz cheers Bruce On Mon, 6 Jul 2015, von Polier, Georg wrote: Dear FreeSurfers, I need to convert an FS-volume back in scanner space (to calculate overlap with manual segmentation). I went through the wiki-entry and converted to native space with mir_vol2vol, however the coordinates are still RAS and not identical with the original Scanner-coordinates. I tried mri_convert, however, the results were not as good as mri_vol2vol (apparently need trilinear interpolation, not available in mri_convert). Any help would be greatly appreciated, Cheers, Georg ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail. ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
[Freesurfer] convert FS volume in scanner space
Dear FreeSurfers, I need to convert an FS-volume back in scanner space (to calculate overlap with manual segmentation). I went through the wiki-entry and converted to native space with mir_vol2vol, however the coordinates are still RAS and not identical with the original Scanner-coordinates. I tried mri_convert, however, the results were not as good as mri_vol2vol (apparently need trilinear interpolation, not available in mri_convert). Any help would be greatly appreciated, Cheers, Georg ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
[Freesurfer] Hippocampal subfields
Dear FreeSurfers, I am just finishing a paper using output from Hippocampal subfields and recognized that the current version has been deprecated. I am interested in volumes of the whole hippocampus. Are these volumes still valid and can you estimate when a new version will be available? Can you provide maybe a brief comment on the decision to deprecate the current version? Best regards, Georg ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
[Freesurfer] freeview and wm-segmentation errors
Hi, using FS 5.1 now, I encountered two problems: 1. when I try to start freeview on my personal laptop, this errors appears: [georg@localhost ~]$ freeview freeview.bin: error while loading shared libraries: libXss.so.1: cannot open shared object file: No such file or directory However, libXss.so.1 is installed on my system running linux (Fedora 15). Do you habe any idea? 2. Running recon-all with -all flag on a cluster (Linux CentOs 5.6) the stream finishes with errors repeatedly showing this message: #@# WM Segmentation Tue May 31 12:36:00 CEST 2011 mri_binarize --i wm.mgz --min 255 --max 255 --o wm255.mgz --count wm255.txt znzTAGskip: tag=544043625, failed to calloc 1952413696 bytes! $Id: mri_binarize.c,v 1.26.2.1 2011/04/08 15:40:50 greve Exp $ cwd /rwthfs/rz/cluster/work/gg855262/subjects_AN/4550/mri cmdline mri_binarize --i wm.mgz --min 255 --max 255 --o wm255.mgz --count wm255.txt sysname Linux input wm.mgz frame 0 nErode3d 0 nErode2d 0 output wm255.mgz Binarizing based on threshold min255 max255 binval1 binvalnot 0 Cannot allocate memory recon-all -s 4550 exited with ERRORS at Tue May 31 12:36:01 CEST 2011 -- | Execution of Batch-Request stopped at Tue May 31 12:36:01 CEST 2011 | peak memory value: 2.386G | real time used: 04:58:38 I did allocate 4 GB of memory and did not face this error when using FS 5.0 or FS 5.1-beta-release with same settings. Any recommendations? Thanks, Georg ___ Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer The information in this e-mail is intended only for the person to whom it is addressed. If you believe this e-mail was sent to you in error and the e-mail contains patient information, please contact the Partners Compliance HelpLine at http://www.partners.org/complianceline . If the e-mail was sent to you in error but does not contain patient information, please contact the sender and properly dispose of the e-mail.
