Comments for Revising MRI http://www.revisemri.com/blog For your short relaxation times. Wed, 21 Dec 2011 18:26:50 +0000 hourly 1 http://wordpress.org/?v=3.3.1 Comment on Fat Suppression by fMRIguy http://www.revisemri.com/blog/2010/fat-suppression/comment-page-1/#comment-74478 fMRIguy Wed, 21 Dec 2011 18:26:50 +0000 http://www.revisemri.com/blog/?p=323#comment-74478 I don't know if I agree that fat saturation is mandatory for effective fMRI studies. It is true that fat will be incorrectly mapped spatially. However, we need to remember that signal variance through time is the most critical feature to preserve in fMRI data, and fat will contribute little to this. The argument that fat signal will decrease your sensitivity to true activation, and result in more false positives is only true if (1) fat increases the level of noise in the voxels in which it is mapped on top of brain tissue, or (2) fat exhibits task related variance. The first is possible, but the contribution of fat to noise through time is probably small considering its relatively poor vascularization. The second is highly unlikely, and can be virtually eliminated from possibility with appropriate fMRI task designs. I don’t know if I agree that fat saturation is mandatory for effective fMRI studies. It is true that fat will be incorrectly mapped spatially. However, we need to remember that signal variance through time is the most critical feature to preserve in fMRI data, and fat will contribute little to this. The argument that fat signal will decrease your sensitivity to true activation, and result in more false positives is only true if (1) fat increases the level of noise in the voxels in which it is mapped on top of brain tissue, or (2) fat exhibits task related variance. The first is possible, but the contribution of fat to noise through time is probably small considering its relatively poor vascularization. The second is highly unlikely, and can be virtually eliminated from possibility with appropriate fMRI task designs.

]]> Comment on Use MRI for: Bees by Jacks of Science » Blog Archive » Bee Brain Boat http://www.revisemri.com/blog/2006/use-mri-for-bees/comment-page-1/#comment-73736 Jacks of Science » Blog Archive » Bee Brain Boat Sun, 27 Nov 2011 18:42:57 +0000 http://www.revisemri.com/blog/2006/use-mri-for-bees/#comment-73736 [...] this at Revising MRI blog while googling Fourier tranforms. Nice find! Posted by Chris Filed in Biology No Comments [...] [...] this at Revising MRI blog while googling Fourier tranforms. Nice find! Posted by Chris Filed in Biology No Comments [...]

]]> Comment on Fat Suppression by practiCalfMRI http://www.revisemri.com/blog/2010/fat-suppression/comment-page-1/#comment-73483 practiCalfMRI Wed, 16 Nov 2011 00:55:14 +0000 http://www.revisemri.com/blog/?p=323#comment-73483 Re. Fat-Sat for fMRI? Yes, it's imperative that you have some sort of fat suppression (or avoidance) scheme with EPI for fMRI. You might be tempted to disable fatsat to save some time per slice, thus increasing the number of slices/TR. However, because the fat protons are always several hundred Hz off resonance from water, the N/2 ghosts from the fat will be VERY intense without fat signal suppression. If you could ensure that these intense ghosts don't overlap brain signal then you would be okay, but if your image field-of-view is typical (below 24x24 cm) then overlap will be unavoidable, and you will have low statistical power in overlap regions (and probably a high number of false positive "activations" in these regions). Re. Fat-Sat for fMRI?

Yes, it’s imperative that you have some sort of fat suppression (or avoidance) scheme with EPI for fMRI. You might be tempted to disable fatsat to save some time per slice, thus increasing the number of slices/TR. However, because the fat protons are always several hundred Hz off resonance from water, the N/2 ghosts from the fat will be VERY intense without fat signal suppression. If you could ensure that these intense ghosts don’t overlap brain signal then you would be okay, but if your image field-of-view is typical (below 24×24 cm) then overlap will be unavoidable, and you will have low statistical power in overlap regions (and probably a high number of false positive “activations” in these regions).

]]> Comment on Calculating confidence intervals by Peggy http://www.revisemri.com/blog/2006/confidence-intervals/comment-page-1/#comment-73056 Peggy Mon, 24 Oct 2011 23:03:07 +0000 http://www.revisemri.com/blog/2006/confidence-intervals/#comment-73056 A 95% cl between between11 and19 means? A 95% cl between between11 and19 means?

]]> Comment on Calculating confidence intervals by jinee http://www.revisemri.com/blog/2006/confidence-intervals/comment-page-1/#comment-72868 jinee Thu, 13 Oct 2011 04:26:05 +0000 http://www.revisemri.com/blog/2006/confidence-intervals/#comment-72868 a surver is conducted on a random sample of 400 americans from the entire American population. Eighty percent (80%) of the respondents agreed with the statement that social secirity benefits should not be taxed and 20% said they should be taxed. At a confidence level of 95%, what is the confidence interval, expressed as 80% +/- confidence interval? a surver is conducted on a random sample of 400 americans from the entire American population. Eighty percent (80%) of the respondents agreed with the statement that social secirity benefits should not be taxed and 20% said they should be taxed. At a confidence level of 95%, what is the confidence interval, expressed as 80% +/- confidence interval?

]]> Comment on Calculating confidence intervals by Preetha RAmakrishnan http://www.revisemri.com/blog/2006/confidence-intervals/comment-page-1/#comment-72844 Preetha RAmakrishnan Wed, 12 Oct 2011 10:37:06 +0000 http://www.revisemri.com/blog/2006/confidence-intervals/#comment-72844 The explanation was clear and useful :-) The explanation was clear and useful :-)

]]> Comment on Diffusion Tensor Imaging Cheat Sheet by Blake http://www.revisemri.com/blog/2008/diffusion-tensor-imaging/comment-page-1/#comment-72709 Blake Mon, 03 Oct 2011 21:18:48 +0000 http://www.revisemri.com/blog/?p=52#comment-72709 Dave, Our Radiologist has been looking for information/guidance/lectures on diffusion tensor imaging as used in traumatic brain injury. He needs info on practical protocols (is there a standard procedure?), how are they being interpreted and what data to interpret, is tractography needed or helpful, how many tracts and which ones do you select, FA values compared to controls, do you do a normal brain then use DTI to look for non-apparent injuries? Thanks Dave,
Our Radiologist has been looking for information/guidance/lectures on diffusion tensor imaging as used in traumatic brain injury. He needs info on practical protocols (is there a standard procedure?), how are they being interpreted and what data to interpret, is tractography needed or helpful, how many tracts and which ones do you select, FA values compared to controls, do you do a normal brain then use DTI to look for non-apparent injuries? Thanks

]]> Comment on Calculating confidence intervals by Maria http://www.revisemri.com/blog/2006/confidence-intervals/comment-page-1/#comment-72610 Maria Tue, 27 Sep 2011 15:28:15 +0000 http://www.revisemri.com/blog/2006/confidence-intervals/#comment-72610 How can i calcualte CI for non-parametric tests, e.g. Mann U How can i calcualte CI for non-parametric tests, e.g. Mann U

]]>