The mgh files look like surface overlays judging from their dimensions (163842 x 1). The values are for corresponding surface vertices. I'm not sure if Freesurfer already has a program to do the conversion. But I think you can do it with the following steps:
1. locate the surface file that the mgh will be overlaid with. This is to retrieve the surface vertex coordinates XYZ in tkr space. 2. get the matrix vox2ras-tkr using 'mris_info <surface_file>' 3. for each surface vertex - convert ras-tkr to vox, CRS=inv(vox2ras-tkr)*XYZ - retrieve the overlay value for the corresponding vertex from mgh - write the value to a 3D volume at the computed CRS location
Yujing
From: freesurfer-bounces@nmr.mgh.harvard.edu freesurfer-bounces@nmr.mgh.harvard.edu On Behalf Of Melissa Thalhammer Sent: Tuesday, February 25, 2025 11:08 AM To: Freesurfer@nmr.mgh.harvard.edu Subject: [Freesurfer] Convert 1D mgh back to 3D
External Email - Use Caution
Dear everyone,
I know that this might be not exactly the right place to ask, but I unfortunately did not reach the authors of the publication so far.
I want to work with the evolutionary expansion map by Wei et al. published in Nat Comm (2019; MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://doi.org/10.1038/s41467-019-12764-8https://secure-web.cisco.com/1EOa2cp_XKYtekyLUN3mqm-iZxVh0pAiE-j1_IvELbdew0_5yco5Qn1zkww73evmjFLHXrpMKk3MTc0UftYqqspUWCnEUttTGobh73r0sLrK4hMDTTJSizcKflNtBNqfTyl0NS51B60IwvBsPsrcslfbsVdLCIjbj4BoGoMOpAmQScRod3QGhL3f8cSiycVoBDa3rvVwLbjfJ4-XsD4jOBXwr5JNfjsjXO9oXVAKmzBdtIH4-G_bH4EZEGPXdEjrocq-7nL1xCjZ0-ENwymFoNFI9Z_3QhFJNvcTLZAXatUskI51ibStKq1lUkZqv4qtfc_w46tNETXWjym-pxRMS7evGrmrL6sH1rEYEwycrDo8/https%3A%2F%2Fdoi.org%2F10.1038%2Fs41467-019-12764-8). On their website, they provide a download link leading to two mgh files (attached; MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://www.dutchconnectomelab.nl/wordpress/wp-content/uploads/Wei_et_al_201...https://secure-web.cisco.com/1wjBHK0nX5LW0U-Wy1w4rY2bystAnHOCIGlVB_T3xRlPwIA1GVvz1pnhCXNC-ia3gnO6FjNtV1pHo2kBNR-Trh5WGlM_e_m6scOf3l8S_0o7rH0SA6Zrn1r6TiqnYqPrACKxhJuGwJ5VzTWlt3WfIQwA0cUywAN5JdXGTmA2PoXdu46XvoRQkV8_nCrPu7KGUhpYo6wJ0o66GZlKJGZoGL9GznJ1khVE-gs2j9XBIl2DEvrfc1c4XcHa_nI5VeWJhgH1jhVc1FwdSQl2JpT7sndUQQRIxPnxXpLYPBbj0ZlL5KifRziPz7JhkCH4iA0cs2PrBGr9xDqQgkZIADtx4QIM6r59o8EkYYRxTCM_UMbI/https%3A%2F%2Fwww.dutchconnectomelab.nl%2Fwordpress%2Fwp-content%2Fuploads%2FWei_et_al_2019_NatureComms.zip). I read them with the Matlab load_mgh.m function provided here (MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be https://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/MghFormathttps://secure-web.cisco.com/1-0OpH0qgfeFpbCMfaVWHm-btTsT6A2ZX44WYfXtCEzjf9bUgQx9CtSi1F45Kk1C855ajx5fNyPBZ7ryU16uJTKx-bTTGtGAzgfOzPkgNgbJfxUH5fhP9wskCP-0amwa1EF4H62ViFuPVhpPAyFBRvKMzinmUWXK17KdFpMtzDVp8k3clvsI0Blg78at2fmOmscYyYGwNShEGrqdV0y3pF9waKvMLx6VNYUBPeauFhhBtHW8Y87V5yOPW9n2DoPUjshSly0_6vqjiVTEx2pgU1MFVv9l3DcdZlxcy7Mbut75yrfmZ8mSFEpJDNS3H3Mq9luH-QoQYLFxVkSQMyi1WKlVCtTJsqqYlD4tf52SeGX8/https%3A%2F%2Fsurfer.nmr.mgh.harvard.edu%2Ffswiki%2FFsTutorial%2FMghFormat) and it shows a 163842 x 1 vector (which probably corresponds to fsaverage6 I guess...). I now want to convertthe 1D vector back into 3D. I have come across save_mgh.m but don't know what should be put in as M (4x4 vox2ras transform). Any ideas what is the correct way to do that?
Thank you!
Best, Melissa