External Email - Use Caution
Freesurfer experts,
I have two MRIs for a patient separated by 5 years. I would like to look at progression of atrophy. These MRIs are different, however, firstly in terms of resolution (new one is 1x1x1, old one is 1.05x1.05x1.2), and secondly just in terms of general contrast, as they are different protocols from different scanners.
First I ran Freesurfer 6.0.0 on both MRIs; but the results I obtained didn't make much sense -- when I compared to a group of controls, the thickness values were much lower in the old MRI than in the new one. This could, of course, be due to variance in contrast / type of scanner, etc. Reviewing the surfaces, in both cases (old and new), all surfaces looked very good with no major errors.
I wanted to see how well the surfaces lined up over time -- so I wanted them in the same space. To get this, I co-registered the new MRI to the old one (via SPM), and re-ran recon-all on it. Now, when I compare to a group of controls, the new MRI has much more expected results -- e.g. lower thickness than the old one. Again, surfaces looked very good.
All that changed is the space the MRI was in when I fed it into FS -- some slight variance from voxel resolutions aside, this gave very different results. I looked at the lh.aparc.stats files and computed some percent differences and was pretty surprised by what I saw. For example, GrayVol values were different by 20% in the precentral, thickaverage was off by 17% in the precuneus.
obviously I made an error somewhere that caused this type of variance but I can't find it -- is there any advice for comparing this type of longitudinal data ?
thank you for any help you can provide!
Quentin Funk, PhD Houston Methodist Research Institute 713-363-9003tel:713-363-9003
Houston Methodist. Leading Medicine. Houston Methodist is ranked by U.S. News & World Report as the No. 1 hospital in Texas for patient care and safety and one of the top 20 hospitals in the nation. In addition, we are nationally ranked in 11 specialties, the most in the state, and have been national leaders throughout the COVID-19 pandemic in research, offering innovative treatments and surpassing CDC safety standards. For more than 100 years, Houston Methodist has provided the best — and safest — clinical care, advanced technology and patient experience. That is our promise of leading medicine. houstonmethodist.org twitter.com/MethodistHosp facebook.com/HoustonMethodist ***CONFIDENTIALITY NOTICE*** This e-mail is the property of Houston Methodist and/or its relevant affiliates and may contain restricted and privileged material for the sole use of the intended recipient(s). Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender and delete all copies of the message. Thank you.
When you mapped to the new space, what kind of interpolation did you use?
On 2/11/2021 11:30 AM, Funk, Quentin wrote:
External Email - Use Caution
Freesurfer experts,
I have two MRIs for a patient separated by 5 years. I would like to look at progression of atrophy. These MRIs are different, however, firstly in terms of resolution (new one is 1x1x1, old one is 1.05x1.05x1.2), and secondly just in terms of general contrast, as they are different protocols from different scanners.
First I ran Freesurfer 6.0.0 on both MRIs; but the results I obtained didn't make much sense -- when I compared to a group of controls, the thickness values were much lower in the old MRI than in the new one. This could, of course, be due to variance in contrast / type of scanner, etc. Reviewing the surfaces, in both cases (old and new), all surfaces looked very good with no major errors.
I wanted to see how well the surfaces lined up over time -- so I wanted them in the same space. To get this, I co-registered the new MRI to the old one (via SPM), and re-ran recon-all on it. Now, when I compare to a group of controls, the new MRI has much more expected results -- e.g. lower thickness than the old one. Again, surfaces looked very good.
All that changed is the space the MRI was in when I fed it into FS -- some slight variance from voxel resolutions aside, this gave very different results. I looked at the lh.aparc.stats files and computed some percent differences and was pretty surprised by what I saw. For example, GrayVol values were different by 20% in the precentral, thickaverage was off by 17% in the precuneus.
obviously I made an error somewhere that caused this type of variance but I can't find it -- is there any advice for comparing this type of longitudinal data ?
thank you for any help you can provide!
Quentin Funk, PhD Houston Methodist Research Institute 713-363-9003 tel:713-363-9003
Houston Methodist. Leading Medicine.
Houston Methodist is ranked by /U.S. News & World Report/ as the No. 1 hospital in Texas for patient care and safety and one of the top 20 hospitals in the nation. In addition, we are nationally ranked in 11 specialties, the most in the state, and have been national leaders throughout the COVID-19 pandemic in research, offering innovative treatments and surpassing CDC safety standards. For more than 100 years, Houston Methodist has provided the best — and safest — clinical care, advanced technology and patient experience. That is our promise of leading medicine.
*MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* houstonmethodist.org https://secure-web.cisco.com/145VEyuEZGESbLjkQykTiaGDW94DoZXcAl008ZzIuMrNJQ-vStUW1fRKE4Rpx2vls_Ug3vqvtNnqZi4QlgxBz2-uwHIcFI9YhudpdAYqBSW0eB3bzi2_7arqTzI3qY7SNbKnZX6oZwDXld1FkW3TYwTU5SI0pjG2SEc-t0BmrsMzRwCme_oMb9dlojpSpkZ85nSuExu8-AS3LjFPzx1NxqMeIdEX1OLG5BO6Zf3KDmkxoLGKzVOf4ZqxH9MJIKplHatsRlsrQNTmfkaZcKIkEYQ/https%3A%2F%2Fwww.houstonmethodist.org%2F *MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* twitter.com/MethodistHosp https://secure-web.cisco.com/1RbHfg9ON4Z3rzcA6z14_GOrNmoDV1V3UhESjhPtRKr_LZqYO9STt6ObxTpgpjifjaH_XdXF5PrPhPnOBqzPNd_t13RHopC092pAz20DHHAXMZrfm9sAsSNXotFbH6n-KvvMCeVSClZn_VBxKRwyJTCpt24ZOX646xShN-Bbc1VAC4hJPZEmMIfimeaNjhDV-d1XSLiFr-vfOIIV7WwPy90sUrtDXLYzuC1oZxTC7Yfb3pStPSgbQ_7cICc7-38U1H5r2jbZxWIrKHro9s8DPzQ/https%3A%2F%2Ftwitter.com%2FMethodistHosp
*MailScanner has detected a possible fraud attempt from "secure-web.cisco.com" claiming to be* facebook.com/HoustonMethodist https://secure-web.cisco.com/1LZFy9Nj4FcDX5PZ-MBnmLfNVADbyKkpIZ0uzK_TaKz3yVLWy7d1TQdaMJEoC4JGxBajUe8Jrq0cz0hMWt485ZtygUtI0z3x1ZpWwKxi-EGqfA2P6e9jr0XPtTBmVnahQBXKSM3dcFxI5t7HvDV5EQ7df8TexgKMIJrskj9YHsEzdM115XUz1cSe-ApkibNVH36bN-5srcj1xLOgPeVkPpxcD6a8rhTCTDU6_5YF38vaj78caREQLJe6JTLSGZpGPF1QNsmR8GV93iFi4tJD1dg/https%3A%2F%2Fwww.facebook.com%2Fhoustonmethodist%2F
***CONFIDENTIALITY NOTICE*** This e-mail is the property of Houston Methodist and/or its relevant affiliates and may contain restricted and privileged material for the sole use of the intended recipient(s). Any review, use, distribution or disclosure by others is strictly prohibited. If you are not the intended recipient (or authorized to receive for the recipient), please contact the sender and delete all copies of the message. Thank you.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Douglas N. Greve -
Funk, Quentin