[Mristudio-users] Skull stripping

[Susumu Mori]

Image registration quality is, in the end, judged by our eyeballs.
If there was a quantitative measure to judge if it's working or not
(regardless of head-to-head or brain-to-brain), we would have used such a
measure to achieve the better automated registration.

So, I would say, if the brains are well registered by the head-to-head,
there is nothing wrong.
Head-to-head does fail more than brain-to-brain. So, extra effort for
skull-stripping may worth, but as long as the results look decent,
head-to-head is fine. When you encounter failure, you could pre-register
them by manual landmarks or by skull-stripping. Either way, blind automated
processing is always risky.

It also depends on image. DTI data with long echo time tend to suppress
non-brain tissue. So the signal is dominated by the brain anyway. T1 has
much more intense signals from outside the brain. So, registration may fail
more with T1 for head-to-head.

In our atlas, we offer both non-skull-stripped and skull-stripped images.




On Mon, May 26, 2014 at 10:55 AM, Shaimaa Abdelsattar <shaimaa96 at hotmail.com
> wrote:

> Thank you Dr. Mori,
> I have followed the manual approach described in
> http://lists.mristudio.org/pipermail/mristudio-users/2010/001382.html and
> i achieved good skull stripping results.
> I understood from your reply that if i do not do skull stripping, the
> operation will be named head to head registration rather than brain to
> brain registration and i see that  ICBM-DTI-81 atlas is skull stripped,
> isn't it?, My question is if i do not do skull stripping, shall i do
> registration with another atlas to have more accurate results of brain
> registration?
>
> Thank you for your kind reply
>
> Shaimaa
>
>
>  On May 26, 2014, at 3:12 PM, "Susumu Mori" <smoriw at gmail.com> wrote:
>
> Hi Shaimaa,
>
> If you don't skull strip, you are doing head-to-head registration. If you
> skull-strip, it is brain-to-brain registration.
> You could argue the latter would be more accurate if you are interested in
> the brain.
> However, the latter could be influenced by the quality of the skull-strip
> and it usually requires manual quality control.
>
>
> On Mon, May 26, 2014 at 9:19 AM, Shaimaa Abdelsattar <
> shaimaa96 at hotmail.com> wrote:
>
>> Thank you very much, i have an inquiry,  should this step be done before
>> normalization to brain atlas, or normalization can be done correctly
>> without this step(skull stripping)?
>>
>> Thanks again
>> Shaimaa
>>
>>
>> On May 25, 2014, at 9:39 PM, "Dorian P." <alb.net at gmail.com> wrote:
>>
>> Hi Shaimaa,
>>
>> If you did not get help on this, I think skull stripping helps. Those are
>> voxels that we don't really need to have a part in the calculation (and
>> they are not just part of the skull but inferior soft tissues as well).
>>
>> I used to do skull stripping by coregistering the T1 on FA, running
>> segmentation of T1, and creating a mask of CSF and white matter. Then just
>> multiply the mask with the FA. This can even be done automatically with a
>> batch.
>>
>> Dorian
>> TJU
>>
>>
>> 2014-05-22 9:50 GMT-04:00 Shaimaa Abdelsattar <shaimaa96 at hotmail.com>:
>>
>>> Hello experts,
>>> Is skull stripping a must do step before doing normalization?, if so,
>>> what is its importance and how can it be done?
>>>
>>> Thanks
>>> _______________________________________________
>>> mristudio-users mailing list
>>> mristudio-users at mristudio.org
>>> http://lists.mristudio.org/mailman/listinfo/
>>> Unsubscribe, send a blank email to:
>>> mristudio-users-unsubscribe at mristudio.org
>>>
>>
>> _______________________________________________
>> mristudio-users mailing list
>> mristudio-users at mristudio.org
>> http://lists.mristudio.org/mailman/listinfo/
>> Unsubscribe, send a blank email to:
>> mristudio-users-unsubscribe at mristudio.org
>>
>>
>> _______________________________________________
>> mristudio-users mailing list
>> mristudio-users at mristudio.org
>> http://lists.mristudio.org/mailman/listinfo/
>> Unsubscribe, send a blank email to:
>> mristudio-users-unsubscribe at mristudio.org
>>
>>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
> _______________________________________________
> mristudio-users mailing list
> mristudio-users at mristudio.org
> http://lists.mristudio.org/mailman/listinfo/
> Unsubscribe, send a blank email to:
> mristudio-users-unsubscribe at mristudio.org
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mristudio.org/pipermail/mristudio-users/attachments/20140527/27ec1938/attachment.html>