Maybe if you gave a few more details about your experiment, as well as the hardware with which you are acquiring, type of sequence, strength of magnet, method of reconstruction (homebrewed or manufacturer) etc, it will be easier to make recommendations. 

But the short answer, I do believe that Peter does mean TR>3 * T2. 

For the short, there are some experiments in which you would acquire with a TR of ~ 250ms, especially if you were concerned with correcting for physiological noise aliasing etc. So there are valid reasons to go quick, but it is not the typical you'll find in the published neuroscience literature.

As to the 20-45ms. According to how fast (that is how fast the gradients can change) your hardware is, each and every slice takes about 20-45 ms to acquire (this is also dependent on the echo time, plus there is additional overhead in the 10's of ms per slice. Thus if you are taking 40 slices you would need ~ 2-3 seconds. This according to the time to acquire 1 slice x the amount of slices will place the lower limit on the TR you can specify, which will be site specific.

I would suggest that you also contact your local engineer that services your machine or any local physics support to get more specific answers.

Best of luck,

Robert Welsh

-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
Robert C. Welsh, PhD
Research Investigator
Department of Radiology
University of Michigan
(734) - 764 - 2412 (fax)
[log in to unmask]


>>> LeiNa Hua <[log in to unmask]> 11/05/05 2:29 AM >>>
Hi SPMer,hi Peter,hi Robert:
 I also have some questions in my mind:
 1,In Peter's mail,*You want to make sure that TR > 3T2, *is it a slip
?sholud be *TR > 3T1*
 2, In Peter's reply,*This sets a minimum TR ~ 300 ms*. but in robert's
mail,he told me *Since you are only acquiring a single slice you can go
quite short, 250ms or 500ms. *can someone explain to me more details?
 3,I know that TR relate to slice number. In Robert's mail, he said:*Typically
an echo planar (and spiral) acquisition per slice is on order 20-45 ms, *Did
he mean that the TR will increase 40-45ms/ one slice when we keep other
paramter fixed? can someone give me more info about the relation between the
TR and slice number in EPI squence?
 4,Hi peter,I can not acess your paper [Kingsley PB. Signal intensities and
T1 calculations in multiple-echo sequences with imperfect pulses. Concepts
Magn Reson 1999; 11: 29-49.] , can you give me a copy?
 Thanks

 On 11/5/05, Kingsley, Peter <[log in to unmask]> wrote:
>
> Leina:
>
> Robert has given some reasons to acquire several slices and use at least a
> moderate TR of ~1 s. You want to make sure that TR > 3T2, otherwise you may
> get unwanted echoes interfering with your data. This sets a minimum TR ~ 300
> ms. The following arguments consider only SNR, not unwanted echoes or the
> number of slices. This SNR is for averaging multiple acquisitions, but the
> results should be applicable to multiple acquisitions that are processed
> separately instead of being averaged. For more information, see [Kingsley
> PB. Signal intensities and T1 calculations in multiple-echo sequences with
> imperfect pulses. Concepts Magn Reson 1999; 11: 29-49.]
>
> No matter how many slices you choose to acquire, once you have chosen a
> TR, the optimum SNR is given by the Ernst angle, cos(theta) = exp(-TR/T1).
> [Ernst RR, Anderson WA. Application of Fourier transform spectroscopy to
> magnetic resonance. Rev Sci Instr 1966; 37: 93-102.] Some approximate
> results are
>
> TR/T1 :4 2 1 0.5 0.3
> angle (degrees) :89 82 68 53 42
> Alternatively, you can set a flip angle, then choose the TR/T1 that gives
> the best SNR. Some approximate results are
> angle (degrees) :90 60 45 30
> TR/T1 :1.25 0.86 0.77 0.72
> Notice that setting TR/T1 and calculating the optimum theta does not give
> the same result as setting theta and calculating the optimum TR/T1.
>
> If you are considering a range of T1, such as WM ~ 700 and GM ~ 1200, I
> suggest using the longer T1 in the Ernst formula.
>
> You can compare the "average SNR per unit time" for any given pair of
> theta and TR/T1 from the formula
> SNR ~ {sin(theta)[1-exp(-TR/T1)]/[1-cos(theta)*exp(-TR/T1)]}/sqrt(TR)
>
> These formulas apply to a single-pulse (gradient echo) acquisition. For
> spin-echo, see the Concepts Magn Reson paper for more details.
>
> -----
> Peter B. Kingsley, MRI Physicist [log in to unmask]
> Department of Radiology / MRI
> North Shore University Hospital
> 300 Community Drive, Manhasset, NY 11030, USA
> Tel (516) 562-2842 Fax (516) 562-3561
>
> -----Original Message-----
> From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]<[log in to unmask]>]
> On Behalf Of Robert Welsh
> Sent: Thursday, November 03, 2005 9:46 PM
> To: [log in to unmask]
> Subject: Re: [SPM] about TR
>
>  Leina,
>
> You should look at your experimental design and let that drive your
> choice of TR in part. If you have an event related design where the
> duration of the stimulus is short then you may opt for a shorter TR.
> Since you are only acquiring a single slice you can go quite short,
> 250ms or 500ms. But I would actually advise taking a few more slices,
> very few studies now days get just a single slice, and in fact usually
> the reason to get a single slice is driven by a desire to acquire them
> very rapidly. Typically an echo planar (and spiral) acquisition per
> slice is on order 20-45 ms. If you don't need to acquire quickly then
> get move coverage. The coverage is especially good for any realignment
> and coregistrations etc.
>
> In much of cognitive neuroscience literature experiments typically have
> a TR or 2 or 3 seconds (state of the art now days) which 30-40 slices.
>
> You should match you flip angle according to the TR. Short TR, less of a
> flip, as Peter has indicated.
>
> -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
> Robert C. Welsh, PhD
> Research Investigator
> Department of Radiology
> University of Michigan
> (734) - 764 - 2412 (fax)
> [log in to unmask]
>
> >>> LeiNa Hua <[log in to unmask]> 11/03/05 8:35 PM >>>
> Hi SPMer,hi Peter:
> When we want to aquire single slice,How should we choose the TR? It
> also
> need TR ~3-4T1?
> Thanks
> Leina
>
>  On 11/3/05, Kingsley, Peter <[log in to unmask]> wrote:
> >
> > It depends how long. With 90-degree pulses, "average" SNR per unit
> time is
> > best when TR/T1 ~ 1.25, so you certainly don't want to use TR <
> 1.25*T1unless you use a lower flip angle. At
> > 1.5 T, T1 is ~700 ms for white matter and ~1200 ms for gray matter,
> and
> > these numbers increase by 10-20% at 3 T. Therefore, for gray matter,
> TR
> > should be at least 1.5 s at 1.5 T and 2 s at 3 T. Beyond 4-6 seconds,
> you
> > will not gain much SNR because you are already ~3-4x T1 for gray
> matter, and
> > therefore almost fully relaxed. A longer TR allows more slices, but I
> see no
> > advantage in setting TR much longer than needed, so TR usually is
> chosen to
> > be between 1.5 s and 5 s.
> >
> > -----
> > Peter B. Kingsley, MRI Physicist [log in to unmask]
> > Tel (516) 562-2842 Fax (516) 562-3561
> >
> >
> > ------------------------------
> > *From:* SPM (Statistical Parametric Mapping)
> [mailto:[log in to unmask] <[log in to unmask]>] *On
> > Behalf Of *chin wei
> > *Sent:* Thursday, November 03, 2005 6:01 AM
> > *To:* [log in to unmask]
> > *Subject:* [SPM] about TR
> >
> > hi all;
> > Who can tell me what is the advantages and disadvantages of a long TR
> > time in EPI scan?
> > thanks in advance
> >
> _____________________________________________________________________
>
>  The information contained in this electronic e-mail transmission
> and any attachments are intended only for the use of the individual
> or entity to whom or to which it is addressed, and may contain
> information that is privileged, confidential and exempt from
> disclosure under applicable law. If the reader of this communication
> is not the intended recipient, or the employee or agent responsible
> for delivering this communication to the intended recipient, you
> are hereby notified that any dissemination, distribution, copying
> or disclosure of this communication and any attachment is strictly
> prohibited. If you have received this transmission in error, please
> notify the sender immediately by telephone and electronic mail,
> and delete the original communication and any attachment from any
> computer, server or other electronic recording or storage device
> or medium. Receipt by anyone other than the intended recipient is
> not a waiver of any attorney-client, physician-patient or other
> privilege. Thank you.
>



**********************************************************

Electronic Mail is not secure, may not be read every day, and should not be used for urgent or sensitive issues.