Back to PLS Helppercentage signal change
Posted on 03/28/11 12:33:42
Number of posts: 16
Hi PLSusers,
I am trying to get percentage signal change for two groups (patients group and control group).
I have a coordinate -44 36 16 (brodmann 46), when I typed in this coordinate under XYZ(mm) box, it gave me invalid number. I just want to know what the reason behind it since i got this coordinate from PLS.
In the analysis, I ignored the invalid number and went to get the signal change( go to window --> response function plot), and I got the pictures of signal change. I just wonder if these pictures are still the signal change with the the coordinate -44 36 16?
Can anyone let me know?
Thank you in advance.
bye
Replies:
Untitled Post
I'm Online
Posted on 03/28/11 13:25:40
Number of posts: 229
I'm wondering if you typed in the voxel coordinate (in pixel #s) rather than in mm??
nancy
nancy
% signal change
Posted on 03/28/11 21:30:30
Number of posts: 16
THank you for your reply. I am pretty sure I put the x y z in mm. Here is what I have done.
I got the cluster report by clicking report and create report. In the cluster report, the order is cluster number lag ; x y z; and x (mm) y (mm) z (mm), etc. I input the number of x (mm) y (mm) z (mm) from the report to xyz (mm) box. I do not think I put voxel number in xyz(mm)box.
If i did not do it wrong, then what could be the possible reasons? Can anyone let me know?
Thank you.
bye
Untitled Post
I'm Online
Posted on 03/29/11 11:03:58
Number of posts: 291
in order to duplicate your problem, please let me know where is your data, and send me the link (please do not attach anything in email).
Untitled Post
Posted on 03/29/11 11:42:03
Number of posts: 16
quote:
in order to duplicate your problem, please let me know where is your data, and send me the link (please do not attach anything in email).
Hi Jimmy,
We sort of figure out how to find the % signal change with that coordinate (-44 36 16). Instead of inputting the coordinate xyz(mm) in axial view. We went to 3D-cardinal view and input the coordinate and no error message shown. But if we input the coordinate from axial view, we still get the error invalid coordinate. We are not sure what the reason is.
the matfile could be found at rri_disks/irene/mah_lab/bing/mean_centered_baseline/mean_centered_nw_incl_null_fMRIresult.mat
the cluster report is in excel file and can be found at rri_disks/irene/mah_lab/bing/mean_centered_baseline/images_mean_centeredPLS/LV1_regions_mean_centered_cluster_incl_baseline_nw_fMRIcluster.xls
The cluster -44 36 16 could be found from LV1 with lag 2 brodmann area 46
thanks
bye
in order to duplicate your problem, please let me know where is your data, and send me the link (please do not attach anything in email).
Untitled Post
I'm Online
Posted on 03/29/11 12:09:21
Number of posts: 291
this is because you can only click (or type) the coordinate to the slice that is shown. change step (first box above) to 1 or 22, and slice 23 (16mm) will be shown, and you will not get error message.
Untitled Post
Posted on 03/31/11 09:31:01
Number of posts: 16
quote:
this is because you can only click (or type) the coordinate to the slice that is shown. change step (first box above) to 1 or 22, and slice 23 (16mm) will be shown, and you will not get error message.
Hi Jimmy,
Thank you for your reply. I did what you have suggested and found the coordinate. I have one more question. How do we know which slice the coordinate will be at. In this case it is at slice 23. But what if the coordinate changes to another one, and how do i find out what slice number has that coordinate. Could you let me know how you find out slice 23 for coordinate -44 36 16?
Thank you.
bing
this is because you can only click (or type) the coordinate to the slice that is shown. change step (first box above) to 1 or 22, and slice 23 (16mm) will be shown, and you will not get error message.
Untitled Post
I'm Online
Posted on 03/31/11 09:44:26
Number of posts: 229
quote:
Hi Jimmy, Thank you for your reply. I did what you have suggested and found the coordinate. I have one more question. How do we know which slice the coordinate will be at. In this case it is at slice 23. But what if the coordinate changes to another one, and how do i find out what slice number has that coordinate. Could you let me know how you find out slice 23 for coordinate -44 36 16? Thank you. bing
Bing Ye....Hi Jimmy, Thank you for your reply. I did what you have suggested and found the coordinate. I have one more question. How do we know which slice the coordinate will be at. In this case it is at slice 23. But what if the coordinate changes to another one, and how do i find out what slice number has that coordinate. Could you let me know how you find out slice 23 for coordinate -44 36 16? Thank you. bing
For any template you choose, the coordinates for any given subject will always match the template...
So if all of your images are registered to one template, and they all have the same origin, there is no possibility of coordinates 'moving' to a different slice.
If you find that coordinates move, you've done something wrong.
Untitled Post
Posted on 03/31/11 10:18:52
Number of posts: 16
quote:
Bing Ye....
For any template you choose, the coordinates for any given subject will always match the template...
So if all of your images are registered to one template, and they all have the same origin, there is no possibility of coordinates 'moving' to a different slice.
If you find that coordinates move, you've done something wrong.
Nancy,
I tried first without moving the slice number, but I cannot find the coordinate (-44 36 16) at saggital view. But when i change the slice number to 23 and I found the coordinate at saggital view. When I chose 3D-Cardinal plane view under window. I could find the coordinate (-44 36 16) without 'moving' the coordinates to a different slice. I am jsut not sure what the reason is and I am not sure what I could have done wrong.
Thank you for your reply.
bing
Bing Ye....
For any template you choose, the coordinates for any given subject will always match the template...
So if all of your images are registered to one template, and they all have the same origin, there is no possibility of coordinates 'moving' to a different slice.
If you find that coordinates move, you've done something wrong.
Untitled Post
I'm Online
Posted on 03/31/11 12:15:30
Number of posts: 229
Hi..
it sounds like either Jimmy or I should look at your dataset to see what's going on...
there should not be any problem with what you are trying to do...
post the datamat & results files on a server and let us know where to get them
nancy
it sounds like either Jimmy or I should look at your dataset to see what's going on...
there should not be any problem with what you are trying to do...
post the datamat & results files on a server and let us know where to get them
nancy
Untitled Post
I'm Online
Posted on 04/01/11 11:49:28
Number of posts: 291
It is the word "moving the slice number" that confused us.
As I said, you can only click (or type) the coordinate to the slice that is shown. If it is not shown, you have to change the "step" in the first box in the result window.
If you don't know which slice in voxel corresponding to the slice in mm, you can use the following simple method:
1. Load any of your scan image. In your case: nii = load_nii('/rri_disks/irene/mah_lab/linda/new_data/hvs/older/dk814/preprocessing/snra_dk814run1.nii');
2. Open the view window: view_nii(nii);
3. Change "Axes Unit" from voxel to millimeter.
4. Type your millimeter coordinate in "[x y z] at crosshair" field.
5. Change "Axes Unit" back from millimeter to voxel.
Now you will see that your coordinate in voxel is: 12 41 23
The last number 23 is the slice that you are looking for.
As I said, you can only click (or type) the coordinate to the slice that is shown. If it is not shown, you have to change the "step" in the first box in the result window.
If you don't know which slice in voxel corresponding to the slice in mm, you can use the following simple method:
1. Load any of your scan image. In your case: nii = load_nii('/rri_disks/irene/mah_lab/linda/new_data/hvs/older/dk814/preprocessing/snra_dk814run1.nii');
2. Open the view window: view_nii(nii);
3. Change "Axes Unit" from voxel to millimeter.
4. Type your millimeter coordinate in "[x y z] at crosshair" field.
5. Change "Axes Unit" back from millimeter to voxel.
Now you will see that your coordinate in voxel is: 12 41 23
The last number 23 is the slice that you are looking for.
Login to reply to this topic.
