 |
 |
HI, Vin: No sure if I¡¯m right, I guess the 0 for the outer ROI on trace image is due to the brain mask used during image processing to create the trace image. Best Jiansong On 9/8/08 11:00 AM, "vin" <[log in to unmask]> wrote: > Hi Martin, > > I am just curious to know, why trace weighted images ? why not others ? > > I just looked SNR using ROIs, and I found SNR b0= 128.19 and on diffusion > image 16.5, and Trace image is not so good to draw the outer ROI, it almost > shows 0 mean/ 0 SD, so when I come bit at the border of brain, than it shows > SNR 9.3. > > Is there any desciption that, how much should be SNR on trace weighted images > ? > > Best, Vin > > > > On Sat, Sep 6, 2008 at 8:24 AM, Martin Kavec <[log in to unmask]> wrote: >> Hi, >> >> I would do ROI, as Vin says, but on a calculated trace-weighted images. By >> this you would avoid a problem of orientation dependent contrast in >> individual DTI images, which could compromise your SNR values. >> >> Best, >> >> Martin >> >> >> On Sat, Sep 6, 2008 at 1:03 AM, vin <[log in to unmask]> wrote: >>> If you are in so much hurry to check SNR over the weekend, than you can use >>> following tip. may be some experienced user will write name of any tool >>> regarding this soon. otherwise you can write one. >>> >>> So, if you have siemens scanner, than in the syngo program under tools you >>> will find possibility to draw a ROI and see the mean / SD of image. Now >>> what you need to do is, to draw two ROIs, one in the center of image and >>> other one outside from the image, but not out side from the image space. >>> >>> now SNR= mean (center)/ sd (outside).. >>> >>> there are several other things in respect to SNR that also matters, but in >>> short I check with this way. >>> >>> comments from other users are welcome on this procedure. >>> >>> vin >>> >>> >>> >>> On Fri, Sep 5, 2008 at 5:25 PM, Kochunov, Peter <[log in to unmask]> >>> wrote: >>>> In my experience FA values are quite sensitive to SNR levels. From my >>>> observations, higher SNR, produced by my directions or by a better coil, >>>> result in a uniform increases in FA. This sort of make sense since FA is >>>> basically a standard deviation-like measurement and is supposed to change >>>> with SNR. >>>> pk >>>> >>>> ________________________________ >>>> >>>> From: FSL - FMRIB's Software Library on behalf of Jiansong Xu User >>>> Sent: Fri 9/5/2008 10:19 AM >>>> To: [log in to unmask] >>>> Subject: [FSL] DTI signal noise ratio >>>> >>>> >>>> >>>> Dear All: >>>> >>>> I'm wondering if any one has tools for assessing the signal/noise ratio of >>>> DTI raw images. >>>> >>>> A recent publication " Diffusion tensor imaging at low SNR: nonmonotonic >>>> behaviors of tensor contrasts. Landman et al. Magnetic Resonance Imaging 26 >>>> (2008) 790¡©800" reported that the FA value was affected by signal/noise >>>> ratio. So, if I want to compare the FA between patients and control >>>> subjects, I need to assess if the signal/noise ratio of two groups were >>>> match. >>>> >>>> Any help is appreciated. >>>> >>>> >>>> Best >>>> >>>> -- >>>> Jiansong Xu, M.D., Ph.D. >>>> Associate Research Scientist >>>> Dept. of Psychiatry >>>> Yale Medical College >>>> 34 Park St., Room S-100 >>>> New Haven, CT 06519 >>> >> > >