Signal-to-noise ratio = mean / standard deviation
On Tue, 18 Mar 2014, Yuko Yotsumoto wrote:
Anastasia,
Thank you so much for your reply.
So I should calculate the mean and s.d. of the values of the low-b within the white matter region for each subject, and compare those numbers between groups. Is it correct to say that the higher mean values indicate larger SNR?
Yuko
On Mar 18, 2014, at 12:53 AM, Anastasia Yendiki wrote:
Hi Yuko - You can use the low-b images and compute the SNR of those images inside a mask of the white matter. Because dt_recon computes a registration from the diffusion to the structural data, you can use that registration to map the white matter label from the freesurfer segmentation (aseg) to the diffusion data. Then calculate the mean and standard deviation of the values of the low-b images inside that white matter mask.
Hope this helps, a.y
On Mon, 17 Mar 2014, Yuko Yotsumoto wrote:
Hello Freesurfers, I have DTI data for two groups of subjects, and were asked SNR for each group. I did ROI analysis on FA values, therefore, I do not have the averaged DTI data. I am wondering if there is any way to compute SNR from dt_recon output. It doesn't have to be the ROI specific SNR values, I just need SNR for the whole DTI data for each group. Any suggestions will be appreciated! Yuko Yotsumoto
Yuko Yotsumoto, Ph. D. Associate Professor The University of Tokyo, Department of Life Sciences Bldg #2, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902 Japan phone : (+81)-3-5454-4332 E-mail: cyuko@mail.ecc.u-tokyo.ac.jp
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.
Hi Anastasia and freesurfers,
I have more questions regarding SNR on DTI data.
I masked the low-b images by the white matter label (aseg), then calculated the SNR ( mean / standard deviation within the label). The mean values ranged for 250-350 across subjects, and the sd ranged for 60-140 across subjects. As a result, I got the SNR ranged for 2.5-4.0 ish.
When I compared these values with the ones reported in other studies, I realized that my values are much smaller.
http://www.ncbi.nlm.nih.gov/pubmed/21591016 http://www.scirp.org/journal/PaperInformation.aspx?paperID=370#.U0JSS8dLFAI
They reported SNR to be around 20-50. It looks like they used different calculations, so probably I should not make direct comparisons, but still the difference worries me.
My questions are: (1) Do our values look valid to you? (2) Are there any article that reported SNR as mean low-b / sd in the white matter aseg mask?
Thank you in advance! Yuko Yotsumoto
Mar 18, 2014 1:26 PM、Anastasia Yendiki ayendiki@nmr.mgh.harvard.edu のメール：
Signal-to-noise ratio = mean / standard deviation
On Tue, 18 Mar 2014, Yuko Yotsumoto wrote:
Anastasia,
Thank you so much for your reply.
So I should calculate the mean and s.d. of the values of the low-b within the white matter region for each subject, and compare those numbers between groups. Is it correct to say that the higher mean values indicate larger SNR?
Yuko
On Mar 18, 2014, at 12:53 AM, Anastasia Yendiki wrote:
Hi Yuko - You can use the low-b images and compute the SNR of those images inside a mask of the white matter. Because dt_recon computes a registration from the diffusion to the structural data, you can use that registration to map the white matter label from the freesurfer segmentation (aseg) to the diffusion data. Then calculate the mean and standard deviation of the values of the low-b images inside that white matter mask.
Hope this helps, a.y
On Mon, 17 Mar 2014, Yuko Yotsumoto wrote:
Hello Freesurfers, I have DTI data for two groups of subjects, and were asked SNR for each group. I did ROI analysis on FA values, therefore, I do not have the averaged DTI data. I am wondering if there is any way to compute SNR from dt_recon output. It doesn't have to be the ROI specific SNR values, I just need SNR for the whole DTI data for each group. Any suggestions will be appreciated! Yuko Yotsumoto
Yuko Yotsumoto, Ph. D. Associate Professor The University of Tokyo, Department of Life Sciences Bldg #2, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902 Japan phone : (+81)-3-5454-4332 E-mail: cyuko@mail.ecc.u-tokyo.ac.jp
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.
Hi Yuko - As long as you define in your paper how you calculated SNR, you'll be fine reporting these values.
a.y
On Mon, 7 Apr 2014, Yuko Yotsumoto wrote:
Hi Anastasia and freesurfers,
I have more questions regarding SNR on DTI data.
I masked the low-b images by the white matter label (aseg), then calculated the SNR ( mean / standard deviation within the label). The mean values ranged for 250-350 across subjects, and the sd ranged for 60-140 across subjects. As a result, I got the SNR ranged for 2.5-4.0 ish.
When I compared these values with the ones reported in other studies, I realized that my values are much smaller.
http://www.ncbi.nlm.nih.gov/pubmed/21591016 http://www.scirp.org/journal/PaperInformation.aspx?paperID=370#.U0JSS8dLFAI
They reported SNR to be around 20-50. It looks like they used different calculations, so probably I should not make direct comparisons, but still the difference worries me.
My questions are: (1) Do our values look valid to you? (2) Are there any article that reported SNR as mean low-b / sd in the white matter aseg mask?
Thank you in advance! Yuko Yotsumoto
Mar 18, 2014 1:26 PM、Anastasia Yendiki ayendiki@nmr.mgh.harvard.edu のメール：
Signal-to-noise ratio = mean / standard deviation
On Tue, 18 Mar 2014, Yuko Yotsumoto wrote:
Anastasia,
Thank you so much for your reply.
So I should calculate the mean and s.d. of the values of the low-b within the white matter region for each subject, and compare those numbers between groups. Is it correct to say that the higher mean values indicate larger SNR?
Yuko
On Mar 18, 2014, at 12:53 AM, Anastasia Yendiki wrote:
Hi Yuko - You can use the low-b images and compute the SNR of those images inside a mask of the white matter. Because dt_recon computes a registration from the diffusion to the structural data, you can use that registration to map the white matter label from the freesurfer segmentation (aseg) to the diffusion data. Then calculate the mean and standard deviation of the values of the low-b images inside that white matter mask.
Hope this helps, a.y
On Mon, 17 Mar 2014, Yuko Yotsumoto wrote:
Hello Freesurfers, I have DTI data for two groups of subjects, and were asked SNR for each group. I did ROI analysis on FA values, therefore, I do not have the averaged DTI data. I am wondering if there is any way to compute SNR from dt_recon output. It doesn't have to be the ROI specific SNR values, I just need SNR for the whole DTI data for each group. Any suggestions will be appreciated! Yuko Yotsumoto
Yuko Yotsumoto, Ph. D. Associate Professor The University of Tokyo, Department of Life Sciences Bldg #2, 3-8-1 Komaba, Meguro-ku, Tokyo 153-8902 Japan phone : (+81)-3-5454-4332 E-mail: cyuko@mail.ecc.u-tokyo.ac.jp
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@nmr.mgh.harvard.edu
-
Anastasia Yendiki -
Yuko Yotsumoto