mrirecon / bart

BART: Toolbox for Computational Magnetic Resonance Imaging
https://mrirecon.github.io/bart/
BSD 3-Clause "New" or "Revised" License
291 stars 161 forks source link

slow ecalib #308

Closed jiaen-liu closed 9 months ago

jiaen-liu commented 1 year ago

I use ecalib to get the sensitivity maps with one slice of 128 channels in the command below. It takes very long. Is there a way to speed up the process? Thanks! [calib emaps] = bart('ecalib -r 10', data); % data is 1x64x48x128

aTrotier commented 1 year ago

Hi Channels number is suppose to be on the third dimensions (readout, phase encoding, partition encoding, channels)

You should permute the dimensions, I suppose you have 48 channels

Aurelien

Le mer. 21 juin 2023 à 18:55, jiaen-liu @.***> a écrit :

I use ecalib to get the sensitivity maps with one slice of 128 channels in the command below. It takes very long. Is there a way to speed up the process? Thanks! [calib emaps] = bart('ecalib -r 10', data); % data is 1x64x48x128

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC5P7O7OTSRHYXUBFLFWZT3XMMRPBANCNFSM6AAAAAAZPBA76Y . You are receiving this because you are subscribed to this thread.Message ID: @.***>

aTrotier commented 1 year ago

My bad, on dim 4. You can check the dimensions expected here : https://github.com/mrirecon/bart/blob/master/src/misc/mri.h

Le mer. 21 juin 2023 à 19:04, Aurélien @.***> a écrit :

Hi Channels number is suppose to be on the third dimensions (readout, phase encoding, partition encoding, channels)

You should permute the dimensions, I suppose you have 48 channels

Aurelien

Le mer. 21 juin 2023 à 18:55, jiaen-liu @.***> a écrit :

I use ecalib to get the sensitivity maps with one slice of 128 channels in the command below. It takes very long. Is there a way to speed up the process? Thanks! [calib emaps] = bart('ecalib -r 10', data); % data is 1x64x48x128

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC5P7O7OTSRHYXUBFLFWZT3XMMRPBANCNFSM6AAAAAAZPBA76Y . You are receiving this because you are subscribed to this thread.Message ID: @.***>

jiaen-liu commented 1 year ago

I followed the example in the bart tutorial where the first dimension is 1. I think that should be fine. It works as if there is only one slice. There are 128 channels in my case.

On Wed, Jun 21, 2023 at 12:05 PM aTrotier @.***> wrote:

Hi Channels number is suppose to be on the third dimensions (readout, phase encoding, partition encoding, channels)

You should permute the dimensions, I suppose you have 48 channels

Aurelien

Le mer. 21 juin 2023 à 18:55, jiaen-liu @.***> a écrit :

I use ecalib to get the sensitivity maps with one slice of 128 channels in the command below. It takes very long. Is there a way to speed up the process? Thanks! [calib emaps] = bart('ecalib -r 10', data); % data is 1x64x48x128

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AC5P7O7OTSRHYXUBFLFWZT3XMMRPBANCNFSM6AAAAAAZPBA76Y>

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308#issuecomment-1601242551, or unsubscribe https://github.com/notifications/unsubscribe-auth/AJ3W3QEZYEHWA2E7N2ZWLQDXMMSVJANCNFSM6AAAAAAZPBA76Y . You are receiving this because you authored the thread.Message ID: @.***>

aTrotier commented 1 year ago

Try reducing the number of coil With coil compression

Le sam. 24 juin 2023 à 00:21, jiaen-liu @.***> a écrit :

I followed the example in the bart tutorial where the first dimension is 1. I think that should be fine. It works as if there is only one slice. There are 128 channels in my case.

On Wed, Jun 21, 2023 at 12:05 PM aTrotier @.***> wrote:

Hi Channels number is suppose to be on the third dimensions (readout, phase encoding, partition encoding, channels)

You should permute the dimensions, I suppose you have 48 channels

Aurelien

Le mer. 21 juin 2023 à 18:55, jiaen-liu @.***> a écrit :

I use ecalib to get the sensitivity maps with one slice of 128 channels in the command below. It takes very long. Is there a way to speed up the process? Thanks! [calib emaps] = bart('ecalib -r 10', data); % data is 1x64x48x128

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308, or unsubscribe <

https://github.com/notifications/unsubscribe-auth/AC5P7O7OTSRHYXUBFLFWZT3XMMRPBANCNFSM6AAAAAAZPBA76Y>

. You are receiving this because you are subscribed to this thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308#issuecomment-1601242551, or unsubscribe < https://github.com/notifications/unsubscribe-auth/AJ3W3QEZYEHWA2E7N2ZWLQDXMMSVJANCNFSM6AAAAAAZPBA76Y>

. You are receiving this because you authored the thread.Message ID: @.***>

— Reply to this email directly, view it on GitHub https://github.com/mrirecon/bart/issues/308#issuecomment-1605053491, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC5P7O7BZUDZNBEGOD5HLDTXMYJHJANCNFSM6AAAAAAZPBA76Y . You are receiving this because you commented.Message ID: @.***>

uecker commented 1 year ago

Please use the mailing list for questions: https://mrirecon.github.io/bart/